A reconstruction of woody vegetation, environment and wood use at Sibudu Cave, South Africa, based on charcoal that is dated between 73 and 72 ka

Abstract

The aim of this paper is to investigate the Sibudu Cave habitat that is dated to the late Pleistocene using archaeological wood charcoal, in order to reconstruct the environment and the activity of the people who interacted with the landscape. Standard anthracology procedures were applied in this qualitative study to analyse charcoal remains from the cave. A representative subset of charcoal remains was subsampled from a larger assemblage and environmental data, as well as evidence of wood use, are interpreted from 72 charcoal types that include 42 types that we identified taxonomically. We highlight that the environment at Sibudu supported a multi-layered Forest with Savanna vegetation based on the presence of many important taxa of these vegetation communities. The wood of the identified taxa has much rot fungi and was also burrowed by pests; however, it is not possible to infer at this stage if the fungi seen here were pathogenic. The presence of fungi is indicative of an environmental setting with high humidity and warm temperatures, such as is optimal for these types of fungi to flourish. Climatic conditions interpreted here agree with previous interpretations that were made from other environmental proxies. These conditions were only intense enough to disturb the microhabitat at Sibudu and did not change the vegetation near the cave. Also noted during the analysis, is that burning wood logs that were infected with brown rot consistently from c.73 to 72 ka probably produced very warm fires.

Introduction

Charcoal remains in archaeological sites are an important source of environmental data that are preserved from the past and provide a means for understanding plant-human-environment interactions throughout prehistory (Esterhuysen and Mitchell, 1996; Asouti and Austin, 2005; Cartwright et al., 2014; Lennox and Wadley, 2019). In southern Africa, studies of archaeological charcoal have contributed to the reconstruction of vegetation history of the subcontinent (Prior and Price Williams, 1985; Scholtz, 1986; Tusenius, 1986; February 1992; Shackleton and Prins, 1993; Esterhuysen and Mitchell, 1996; Cartwright and Parkington, 1997; February, 2000; Allott, 2006; Cartwright et al., 2014; Chikumbirike, 2014; Bamford, 2015; Lennox and Wadley, 2019). The southern African Middle Stone Age record is rich with evidence of the innovative activities that mark important human behavioural evolutionary stages (Lombard et al., 2012). Many Middle Stone Age sites, including Sibudu Cave, contain evidence for behavioural changes through which the transition from one evolutionary milestone to another is understood (Wadley, 2007; Henshilwood et al., 2009; Brown et al., 2012; Lombard et al., 2012). Apart from social factors, one of the things that may have influenced innovation during the Stone Age is the significant change in the environment and consequently the resources on which people depended through time (Deacon, 1992; Ambrose, 1998; Fitzhugh, 2001).

There is a growing need, therefore, to understand the relationship between environmental/climatic drivers and resource fluctuation as well as that between resource fluctuation and human behaviour. Sibudu Cave provides the platform to study the transition from one Stone Age innovation to another due to its’ rich archaeological record that spans many important evolutionary stages. As a result, these stages can be easily investigated in relation to potential global climate-influencing incidents, such as the youngest volcano super-eruption of Mount Toba in Indonesia which took place approximately 74 ka, and might have affected local habitats during the late Pleistocene (Ambrose, 1998). This paper is focused on presenting the results for the analysis of charcoal samples from Sibudu Cave that were recovered from an archaeological layer dated between c. 73 and 72 ka, a period that represents a transition between the pre-Still Bay and Still Bay lithic industries.

Sibudu Cave is located at 29°31′21.50″S, 31° 5′9.20″E in the province of KwaZulu-Natal, South Africa, on a cliff that, today, is 100 m above mean sea level (a.m.s.l.) (Fig. 1). The cave is situated within an intersection of at least three major vegetation communities of southern Africa c. 15 km from the Indian Ocean (Mucina and Rutherford, 2006). The vegetation in this area grows on the soils that are derived from shale and sandstone rocks near the uThongathi River (Wadley and Jacobs, 2004). Contemporary woody vegetation near the cave has been described and species recorded by various botanists (Mucina and Rutherford, 2006). The current vegetation record at Sibudu, however, may be inaccurate due to sugar cane farming near the cave that has been responsible for the clearance of a large portion of indigenous vegetation. Mucina and Rutherford (2006), for example, mapped three vegetation communities near the cave: Savanna, Indian Ocean Coastal Belt and Forest vegetation communities. At present, there are approximately hundreds of species within these communities and more than 100 indigenous woody species in the vicinity of the cave (Allott, 2005).

The excavation of charcoal specimens discussed here, was done between 1998 and 2011 under the direction of Prof. Lyn Wadley (Wadley and Jacobs, 2004, 2006). The deposit was excavated following the natural stratigraphy, however layers that are deeper than 10 cm were arbitrarily split and named respectively after the colour of the dominant matrix of each layer (Wadley and Jacobs, 2006). The stratigraphy of the Middle Stone Age at Sibudu is clearly defined by different colours and contains ash (Wadley and Jacobs, 2004, 2006). Hearths are found in all layers and are generally defined by flat circular ash remains, burnt soil sediments at the bottom and with concentrations of charcoal at the centre (Wadley and Jacobs, 2006). Botanical remains from their deposit show evidence for the collection of plants for many uses (Allott, 2006; Wadley et al., 2011; Lennox and Bamford, 2015; Lennox et al., 2015; Lennox and Wadley, 2019).

Three layers are discussed here from the Sibudu stratigraphy. The Brown Sand (BS) layer is dated, through single grain Optically Stimulated Luminescence (OSL), to 77.2 ± 2.1 ka. The Light Brownish-Grey (LBG) layer (split into sub-layers LBG(1), LBG2, LBG3 and LBG4) is dated between 72.5 ± 2.0 (LBG1) ka and 73.2 ± 2.3 ka (LBG2), and the Reddish-Grey Sand layer (RGS and RGS2) is dated to 70.5 ± 2.0 ka (Fig. 1) (Jacobs et al., 2008). These layers show that the cave was occupied regularly by people during the late Pleistocene and contain charcoal. The charcoal remains discussed in this paper were recovered from LBG1 to LBG3 along with other archaeological artefacts in four squares: B4, C4, B5 and C5, that are measured at 1 m² each (Wadley and Jacobs, 2006). These remains were curated separately, and the charcoal samples were stored temporarily at the Evolutionary Studies Institute, University of the Witwatersrand. Layer LBG contains the pre-Still Bay lithic technology and the light brownish-grey soil sediments of the layer accumulated during a period of low human occupation of the cave compared to the preceding BS layer (Clark, 2019). Volcanic ashes including that of the ~75 ka Toba eruption have not yet been recognised in the Sibudu archaeological deposit; however, such events are temporal markers whose effect can be traced on global climate, ecosystem and maybe even human behaviour. Charcoal remains from Sibudu were, therefore, analysed to understand if the growth of vegetation as well as people's choice of fuelwood was affected by sudden short- or long-term climatic perturbations during the late Pleistocene. The charcoal assemblage is assessed qualitatively for indicators of the environment in which the vegetation grew as well as evidence of wood use with the goal of describing the finer details of the landscape that accompanied the cultural changes that are documented at this time. This reconstruction will attempt to address the potential intensity of the climatic conditions, the quality -and to a lesser extent quantity-of woody vegetation that was burned as well as a new approach for understanding wood use by considering modifications of wood anatomy related to fungal activity. All charcoal samples analysed here were introduced into the cave anthropogenically (Schiegl and Conard, 2006; Goldberg et al., 2009; Bruch et al., 2012).