Through previous works in the early 1930s by C. Arambourg and in the 1960–1970s by the International Omo Research Expedition (IORE) initiated by F. C. Howell, the Omo Group deposits of the Lower Omo Valley provided decisive data on Plio-Pleistocene environmental change and hominid evolution in eastern Africa. Y. Coppens directed the IORE French component with Arambourg, then alone from 1970 to 1976. After 30-year hiatus, the Omo Group Research Expedition reinitiated field work on Shungura Formation deposits aged between 3 Ma and 2 Ma. In 2006 and 2007, renewed methods led to the collection of more than 600 vertebrate specimens with a particularly precise record of contextual data. These specimens include significant hominid remains dated to 2.5 Ma and slightly older. Changes in faunal distributions were also recorded. Additionally, the Shungura Formation archaeological record is reconsidered. These first results are indicative of future advances in the study of biodiversity evolution and its relationship with global and regional environmental changes.
The Omo Group Research Expedition, an international multidisciplinary team, was created in 2006 to reinitiate palaeoanthropological field work on the Omo Group sediments found in the Lower Omo Valley, south-western Ethiopia (
Beyond the scientific significance of its fossils and artefacts discovered during the 20th century, the Omo Group deposits of the Lower Omo Valley, notably the Shungura Formation, feature prominently in the history of research on human evolution. It was one of the first palaeoanthropological research areas known in eastern Africa, and was exploited by a diversity of teams with a correlated diversity of methods, goals, and theoretical frameworks. Guillemot
The French palaeontologist Arambourg was initially drawn to the Lower Omo Valley by fossils collected from there in 1902 during the trans-African expedition of du Bourg de Bozas. Arambourg conducted a field expedition through Kenya and worked mostly on the southern part of the Lower Omo fossiliferous deposits from January to March 1933. He recognized the Plio-Pleistocene age of these localities and broadly described their stratigraphy and sedimentology
If Arambourg led the way to important fossil discoveries in the Lower Omo Valley, the work of F. Clark Howell was decisive in a methodological dimension and had a particularly wide impact on the whole field of palaeoanthropology. His first research in the Omo Valley was conducted in 1959. This short expedition led to the formation of the International Omo Research Expedition (IORE) including American, French, and Kenyan scientists. The multidisciplinary teams (including notably geologists, radiochronologists, palaeontologists, palynologists, archaeologists, anthropologists) conducted fieldwork for a duration of nine years between 1967 and 1976. The IORE work was marked by the ability of Howell to work at the interface of all disciplines related to human evolutionary studies, from tectonics to artefact technology. He also understood fully the potential of newly developed methods, such as radiochronology, and quickly incorporated them. Consequently, the Shungura Formation, sampled at hundreds of localities, was divided into 111 stratigraphic units precisely described
Since the 1980s, palaeoanthropological research experienced considerable advances. New species were discovered from Middle to Early Pliocene
During the 3-2 million year window, significant and crucial events in human evolution took place. They were first described and theorized by Coppens as the “(H)Omo Event”
Second, increasingly old behavioural evidence of significant diversity
Third, an abundant literature proposes that climatic changes taking between 3.0 Ma and 2.0 Ma had a determining affect on hominid evolution
The Omo fossil record, in its chronological span and completeness, can best address these complex issues
Despite early intensive research, the potential of the Shungura formation was by no means exhausted during the 1960s–1970s. This is shown by the discovery of many new localities, important hominid remains
In light of the issues above detailed, the Omo Group Research Expedition has six basic goals.
On this ground, the Omo Group Research Expedition targets particularly the reconstruction of wet habitat ecosystems. Wet habitats concentrate important resources for large segments of tropical faunas, and this no doubt included extinct hominid species. These habitats may have played an important role in hominid dispersal
Field work was conducted for two weeks in July 2006 and two weeks in June 2007 in the Shungura Formation. Access to the formation was made extremely difficult by poor road conditions. These first missions, including 15 to 18 members, focused on palaeontological survey and collection. The following array of methods is used by the Omo Group Research Expedition: controlled group survey on predefined areas according to geological context; ‘crawling’ surveys on localized areas with particular fossil content
All collected specimens are prepared, studied, and stored at the National Museum of Ethiopia located in Addis Ababa. They receive inventory numbers built on the following model:
[locality identifier] [locality number]/[locality sub-number if required]-[specimen number].
Examples: OMO 18/sup-10025; L 293-10001; OMO 333-10003.
Whereas old localities were identified by a range of letters or group of letters (L, P, F, OMO), all new localities are indicated solely by the identifier ‘OMO’ and locality numbering follows from the last IORE ‘OMO’ number (OMO 327). Newly collected specimens from both old and new localities are numbered starting from 10001. This avoids any possible confusion with IORE collections where the highest specimen number reached 9089. Unlike the case of previous French IORE collections, the year of discovery is not included within the number and the specimen numbers are incremented independently from year changes. Taking collection years into account within inventory numbers greatly elongates the numbers for no significant advantage, and has furthermore proved to be a constant source of curatorial error.
All inventory and contextual data are stored in a newly developed relational database built for integrating previous computerized catalogues. This database follows the Revealing Hominid Origins Initiative database template
In 2006–2007, a total of 27 localities were surveyed, all located in the southern part of the type area of the Shungura Formation, essentially the area where the IORE French team worked. In these localities, 639 fossil specimens identifiable to the genus level were collected. A few localities were surveyed in Members D (OMO 119, OMO 120) and G (OMO 310, OMO 323, OMO 330), but the majority of sampled localities were located in Members C and B (
Seven new localities were defined, either to delimit more precisely fossil accumulation areas within larger localities (OMO 329 within OMO 56sup), or to inventory previously uncollected loci. In terms of fossil content conservation and renewal, three categories of localities were identified.
The first category concerns localities heavily collected by the IORE, such as OMO 18, which appeared much depleted compared to the initial content found by the IORE. At OMO 18, surveyed in 2007 by 10 people during three days, first estimates corrected according to sampling methods (by applying 2007 selectiveness to the IORE fossil record) indicate a renewal rate of about 11%. In other words, for a given category of elements from a given family collected by both IORE and Omo Group Research Expedition, the IORE sample is on average 9.1 times larger than that collected by the Omo Group Research Expedition. It is difficult to estimate differences in sampling efficiencies of previous and new field campaigns. If these differences are negligible, the renewal rate indicated here appears relatively important compared to expectations
The second category of locality comprises IORE localities where 2006–2007 surveys found many more fossils than during the whole IORE operations. This was the case for localities immediately adjacent to first category localities, as for OMO 18/sup, located in the same catchment as OMO 18. In 2007, comparable amounts of fossils were collected in both localities (i.e., 63 specimens in OMO 18/sup versus 78 specimens in OMO 18) whereas the IORE found about 35 times more specimens in OMO 18 than in OMO 18/sup.
The third category includes previously non-inventoried localities found outside previous localities. Until now, these new localities are limited in size and do not contain large amount of fossils. These features indicate that, in the surveyed area, the IORE identified most fossil concentrations, but apparently focused its efforts on the richest of these concentrations.
Detailed examination of the collected fauna is currently in progress, but preliminary study of the 2006–2007 specimens enabled a combined faunal list for Member B and Member C (
Cercopithecids are followed in relative abundance by bovids (120 specimens). Tragelaphini are dominant in all members, but Reduncini, infrequent in Members B and C, are abundant in Member G.
Suids are also common (86 specimens) and relatively diverse. Tetraconodontinae are well represented, notably in Member B with
Hippopotamids constitute the fourth most common group (60 specimens). The collected material should allow pinpointing precisely the transition between hexaprotodont and tetraprotodont forms of aff.
Five new hominid specimens were discovered in five different localities of Member C, and can be dated from 2.74 Ma to 2.58 Ma. This proportion of specimens fits expectations based on IORE collections. This sample includes: two upper molar crowns, one complete (OMO 84-10001) and one fragmentary (OMO 18/inf-10022); two complete lower molar crowns (OMO 224-10005 and OMO 329-10015); one edentulous mandible with preserved symphyseal area and partial left and right corpora (OMO 333-10003). The molar cusp pattern is simple, with no or few additional cusps/ids. A preliminary examination excludes the interpretation that this material belongs to new taxa. These teeth can be attributed to two different taxa: the lower molars belong to a form belonging or close to genus
Faunal distributions exhibit some differences between the different members (
A preliminary reassessment of the existing archaeological evidence collected by the IORE and stored at the National Museum of Ethiopia was conducted by A. Delagnes in collaboration with Y. Beyene. This examination included published material from OMO 71, OMO 84, OMO 57, and OMO 123
Re-examination focused on a technological diagnosis of all the series, complemented by a quantitative analysis of the in situ material from the two largest series: OMO 123 (member F, 266 lithic remains) and OMO 84 (Member E, 161 lithic remains). The recorded data includes, for each category of product (flakes, fragments of flakes, undetermined fragments <1 cm, chunks, cores): raw material characterization, cortex extension (recorded by 25% classes) and size attributes. This preliminary analysis confirms the general features of the Omo assemblages, initially documented by Chavaillon
All this implies that a technological reinterpretation of these series should not be conducted without detailed contextual assessment of the sites from which they derive. This assessment should include: (1) taphonomy of archaeological occurrences (impact of fluviatile dynamics on site formation processes, extent of postdepositional processes, i.e. argiliturbation or bioturbation); (2) location and petrographic characteristic of the raw material potential sources; (3) physical and mechanical assessment of the Lower Omo Valley quartz properties with regard to knapping. These aspects, largely neglected in previous works, will guide future archaeological work in the Shungura Formation.
The work of the Omo Group Research Expedition will be intensified during the next field seasons. Paleontological surveys will be extended to the complete sequence of the Omo Group deposits in the Lower Omo Valley (including Usno and Mursi Formations). Revision of the stratigraphy is required for a number of localities in the Shungura Formation
The past and future work of the Omo Group Research Expedition is dedicated to F. Clark Howell. We terribly miss his support, his wisdom, and his friendship.
We wish to thank all participants in the field work, who endured particularly difficult conditions and devoted their energy to its success (Bizuayehu T., Kampiro K., A. Souron, Woganu A., E. Charpentier, M. Guillemot, Beza S., Godana W., Sissai G., Yohannes M., Amare S., Temechach Y., Tesfaye A., Workeneh Z., GebreSelassie S., Lokora K., Lokulan M., Lopeto L., Awake A., Bamlaku T., Dagne G.). We are thankful to the Ethiopian authorities and their staff for their critical support: Authority for Research, Conservation, and Cultural Heritage (ARCCH, dir.: J. HaileMariam); National Museum of Ethiopia (dir.: M. Yilma); Southern Nations, Nationalities, and People Region (SNNPR). Field work greatly benefited from the assistance of the Centre français des études éthiopiennes (CFEE, dir.: F.-X. Fauvelle). We would like to thank the following colleagues for their decisive help and advice: B. Asfaw, M. Brunet, T. D. White, P. Vignaud, H. de Lumley, G. WoldeGabriel. We are very grateful to A. Souron, E. GebreEgziabher, C. Noël, A. Fernandes Bastos, G. Florent, H. T. Mackaye, and A. Likius for their particularly appreciated contributions. The following institutions and research teams provided funding and/or assistance: Fondation Fyssen; Agence nationale pour la recherche (ANR M. Brunet); ministère des Affaires étrangères (sous-direction de l’Archéologie); CNRS Eclipse II (H. Roche, P. Vignaud); IPHEP UMR 6046, univ. Poitiers; CFEE; Middle Awash Research Project; Mission paléoanthropologique francotchadienne;
Spatial location of the Shungura Formation, Lower Omo Valley, Ethiopia. AA: Addis Ababa; star: Shungura Formation.
Field work methods used during the 2006–2007 field seasons, Shungura Formation, Lower Omo Valley, Ethiopia. A: controlled survey and specimen flagging; B: localized excavation; C: ‘crawling’; D: sieving operation.
Spatial location of Member C and Member B localities surveyed in 2006 and 2007 in the southern part of the Shungura Formation type area. Fossil hominid localities are: OMO 18inf, OMO 84, OMO 224, OMO 329, OMO 333. OMO 112 includes four localities: OMO 112/1, OMO 112/2, OMO 112/3, and OMO 112/4. Recent deposits include Middle to Upper Pleistocene and Holocene sediments.
2006–2007 specimen distribution by higher rank taxa in Members B, C, and G.
Faunal list for Member B and C localities sampled by the Omo Group Research Expedition in 2006–2007
Taxa found in Member B only; no a: taxa found in Members B and C.
aTaxons collectés uniquement dans le membre B ; absence de a : taxons collectés dans les membres B et C.
Taxa found in Member C only.
bTaxons collectés uniquement dans le membre C.