Fire and burning in West Africa Holocene savanna palaeoenvironment. Anthropogenic and natural processes in environmental changes.
by
Ballouche, Aziz
Géophen, UMR LETG 6554-CNRS, UFR de Géographie, Université de Caen, BP 5186, F-14032 Caen, France.

For many, it is generally considered that fire is one of nature's most dramatic processes. It is not only vegetation that is affected, but also animals that may be killed and soils that may be degraded, but in some cases may benefit from the new plant growth. On the other hand, fire must have been considered as an important factor in shaping the present savannah landscapes. In most West African savannas, fires currently occur today during the dry season. Fire management and fire behaviour studies, propagation and spatial-temporal dynamic mapping are a main theme in our approach. Human interaction with fire and vegetation occurs at many levels of human population density and cultural development, from subsistence cultures to highly technological societies. The dynamics of these interactions with respect to wild fire are often difficult to understand and identify in the past. In West African savannas we know little about the ways in which changes in human population and culture alter the use of anthropogenic fire and consequently how that dynamic affects geosystem processes and attributes. Fire and burning are not regarded as a “catastrophe” but there are sudden or a short-duration events, that have a significant effect on vegetation dynamic on a local scale and on ecosystem and landscape changes on a regional or global scale. In this paper three levels of analysis are presented: a spatial and dynamic dimension, a historic dimension. Spatial and dynamic value of fire Today most of the West African savannas are cultural landscapes which have been strongly shaped by humans. A systemic analysis of the West-African woodlands/savannas mosaic shows that it’s an interesting example of functioning and development of a complex system of relationships between societies and their environment. Many studies on modern vegetation covers show that their actual physiognomy and species composition are highly influenced by human impact, in particular shifting cultivation, grazing and firewood exploitation. The modern distribution pattern of forest, woodland and savannah in West Africa equally can be explained as resulting from extensive shifting cultivation. Fire is regarded as the major factor of the landscape dynamics, which helps to maintain the woodland/savannah mosaic. Fig. 1: Spatial dimension of fire in inter-tropical Africa in 1999-2000. Fire regimes throughout the West African savannas are altering. Fire pattern heterogeneity, which encompasses a range of different fire regimes over time and space, has been applied by indigenous people in the past but is increasingly being adopted by fire managers today. Fire history and landscape building A starting point is the controversial discussion on the origin of the savannas: is their existence mainly due to natural factors (climate, soils), or should they be considered as anthropogenous, i.e. resulting from human impact and the degradation of former forests? In Africa, fire history runs across geological timescales, from the pre-Quaternary times, into the present day (Boulvert 1990, Schulz and Pomel 1992, Weiss et al. 1996, Bird and Cali 1998, Salzmann 2000, Wooller et al. 2000, Ballouche 2002). The microscopic charcoal content of several Holocene pollen sequences is used to investigate fire history in West Africa during the Holocene. Although fluctuations in charcoal composition are recorded, it is difficult to link them directly to either human-made or natural fires. Perhaps the most significant alteration to fires in the Holocene palaeoenvironment has been the use of fire by humans. After Salzmann (2000), for example, the charcoal particle curve of Lake Tilla (Biu Plateau, Nigeria) supports the assumption that frequent fires constitute a major agent, which maintains the open character of the savannas vegetation, during all the Holocene times. In this site, the constancy of the charcoal curve provides no evidence whether these fires were mainly of natural origin or induced by early hunters, who may have started them long before the onset of the Holocene. We not confirm this interpretation and we present a synthesis of archaeological and palaeoenvironmental data for the Holocene period based on an example of West African settlement from the Dogon Plateau in Mali. An example: fire and anthropogenic impact on landscapes The Ounjougou site complex (Huysecom 1996, 2002, Huysecom et al. 2002), on the Dogon (Bandiagara) Plateau in Mali, is situated around ten kilometres east of the city of Bandiagara, on a perennial river, the Yamé. This complex includes numerous archaeological sites. It presents as a series of gullies cut into a complex succession of Quaternary aeolian, alluvial and colluvial deposits. The 16 meter thick stratigraphic sequence has yielded archaeological material from the Lower Palaeolithic to modern times. The sediments also contain abundant vegetal remains (pollen, leaves, charcoal, wood, seeds, etc.) for which the state of preservation is exceptional for the southern Sahara (Cf. Fig. 2). The results of our research currently permit us to define five principal occupation phases on the basis of chronostratigraphic, archaeological and palaeoenvironmental data (Huysecom et al. in press). (Cf. fig. 3). Fig. 2: Palynofacies with rich charcoal particles (Ounjougou/Mali, 4th millennia BC). Ounjougou - phase 1 (10th – beginning of the 9th millennium BC) After a favourable climatic period, characterised by relatively dense and diversified Palaeolithic occupations, the arid Ogolian begins locally around 23000 years BP and is represented at Ounjougou by a significant depositional and archaeological hiatus. It is not until the Holocene and the return of humid climatic conditions, beginning in the 10th millennium BC, that it is possible to again observe evidence of human occupation. The total absence, at present, of charcoal in the beginning of phase 1 probably indicates the rarity of fires during these initial humid periods. Some charcoal are present at the transition between the 10th and 9th millennia BC, but it is currently impossible to determine if it is of anthropic origin or natural. Ounjougou - phase 2 (8th millennium BC) In a depositional context indicating a strong hydrologic capacity, the fill of a large channel shows finely bedded mud alternating with levels relatively rich in charcoal. This probably indicates a desire to control the environment by the use of fire beginning in phase 2. The surrounding vegetation is savannah, with the presence of Detarium, as well as a gallery forest with Syzygium. Ounjougou - phase 3 (5th-4th millennia BC) The site contains fine deposits dating to the end of the 4th millennium BC, between 3240 and 3100 yr BC, rich in organic material and charcoal. Anthracological analyses indicate gallery forest with Syzygium dominant and the presence of Uapaca on the edges of the river. The interpretation of palynological taxa from vegetal formations of drained soils permits the reconstruction of a large scale mosaic landscape, dominated by savannas with shea butter trees (Vittelaria paradoxa) and Combretaceae, alternating locally with Isoberlinia woodlands, and perhaps even dry forests. Moreover, we note the regularity of generalized fires, probably of human origin, which could explain the open character of the landscape. Fig. 3: Settlement history and palaeoenvironment of Ounjougou (Dogon Plateau, Mali). (after Huysecom et al. 2003) Ounjougou - phase 4 (3rd millennium BC) Analyses of vegetal remains of the phase 4 levels indicate that the banks of the Yamé river sheltered a gallery forest with less southern affinities, including Albizia, Syzygium and Alchornea, situated in a general savannah environment, documented by Sophia charcoal and pollen from grasses, Combretaceae, shea butter tree, etc. Ounjougou - phase 5 (2nd millennium BC) The site of Varves consists of alternating fine grey silty layers, very rich in organic material and archaeological objects, attaining a stratigraphic thickness of around two metres. The sediments indicate a reduction in the capacity of the Yamé and a change in sedimentation conditions. A more or less regular alternating grey charcoal-rich silts and fine sands are deposits in the extension of stagnant waters, evidencing more sporadic flooding. A seasonal function is sometimes recognizable, but it remains difficult to interpret the geometry of the deposits. Pollen and archaeobotanical analyses indicate the importance of herbaceous plants, and more particularly large grasses. On the edge of the Yamé River, the gallery-forest with Syzygium and Alchornea observable in phase 4 persists in unit 5a, with the presence of a bamboo, Oxytenanthera abyssinica. We note in contrast the appearance on the savannah of taxa with more Sahelian affinities. The layers of unit 5b reveal that the gallery-forest continued to the end of the 2nd millennium, while Terminalia glaucescens and Daniellia oliveri were found on the savannah. Globally, the change in vegetation observed between phases 4 and 5 seems to indicate an aridification of the landscape, developing toward a Sudano-Sahelian savannah with extrazonal taxa that persist. The importance of the ashy microremains indicates the frequency of large fires, quite probably anthropogenic. The interdisciplinary approach employed at Ounjougou reveals an interesting example of function and development of a complex system of relationships between human societies and their environment, demonstrating important changes during the Holocene. The coincidence of large amounts of charcoal and archaeobotanical remains (grinding stones, ceramics, wooden artefacts) in these deposits might indicate that fire and burning, in relation with human activities, played an important role in the building of the landscape during this period. Conclusion Fire has been recognised today as an important vegetation ecological factor and has been incorporated into vegetation dynamic models. In all West-African savannas, his impact on vegetation in the past is good documented. But it is not easy to relate palaeoenvironmental data closely to fire frequency and origin because sediment supply, climate or human activities, which are difficult to unravel, can influence fluctuations. Fire must always have been a part of the Holocene landscape dynamics in the west-African savannas. We think that anthropogenic, or human-caused, fire has influenced ecosystem processes for millennia, and at broader scales has been an important determinant of landscape character. However, data on the past show that vegetation landscapes changes, environments dynamics, societies’ history and land use are closely linked. Economic and social influences have produced true cultural landscapes owning patrimonial value. Ballouche A., 2002. Histoire des paysages végétaux et mémoire des sociétés dans les savanes ouest-africaines. Historiens et géographes. 381. 379-388. Bird M.I. and Cali J.A., 1998. A million-year record of fire in sub-Saharan Africa. Nature, 394. 767-776. Boulvert Y., 1990. Avancée ou recul de la forêt centrafricaine: changements climatiques, influence de l'homme et notamment des feux. in: Lanfranchi R. & Schwartz D. (eds.), Paysages quaternaires de l'Afrique centrale atlantique. Paris, ORSTOM: 353-366. Huysecom E., 1996. Découverte récente d'un site stratifié holocène à Oundjougou, Mali: résultat des deux premières missions préliminaires. Nyame Akuma 46. 59-71. Huysecom E., 2002. Palaeoenvironment and human population in West Africa: an international research project in Mali. Antiquity 76. 335-336. Huysecom E., Ballouche A., Boeda E Cappa, L. Cissé, A. Dembélé, A. Gallay, D. Konaté, A. Mayor, S. Ozainne, F. Raeli, M. Rasse, A. Robert, C. Robion, K. Sanogo, S. Soriano, O. Sow and S. Stokes, 2002. Cinquième campagne de recherches à Ounjougou (Mali). Swiss-Liechtenstein Foundation for Archaeological Research Abroad, Jahresbericht 2001. Zürich: Tamedia, 55-113. Huysecom E., Oainne S., Raelif F., Ballouche A., Rassem M. and Stokes S., 2003. Ounjougou (Mali): A history of Holocene settlement. Antiquity, in press. Salzmann U., 2000. Are savannas degraded forests? - A Holocene pollen record from the Sudanian zone of NE-Nigeria. Vegetation History and Archaeobotany 9: 1–15. Scott L., 2002. Microscopic charcoal in sediments and Late Quaternary fire history of the grassland and savanna regions in South Africa. Jour. Quaternary Science, 17 (1): 77-86. Schulz E.and Pomel S., 1992. Die anthropogene Entstehung des Sahel. Würzburg. Geogr. Arbeiten, 84: 263-288. Weiss K.F., Goldammer J.G., Clark J.S., Livingstone D.A. and Andreae M.O. (1996). Reconstruction of Prehistoric Fire Regimes in East Africa by Lake Sediment Analysis. In: Levin J.S., editor. Biomass burning and Global change. 1, Cambridge (MA): MIT Press: 545. Wooller M.J., Street-Perrot F.A., Agnew A.D.Q., 2000. Late Quaternary fires and grassland palaeoecology of Mount Kenya, East Africa: evidence from charred grass cuticles in lake sediments. Palaeogeography, Palaeoclimatology, Palaeoecology. 164: 207–230.

Date received: November 14, 2003