Tracing the source of sediment in Blombos Cave, South Africa, during the period of Middle Stone Age occupation.

Overview
This research project was developed in collaboration with the above participants. The project is designed as part of the Center for Early Sapiens Behavior (SapienCE) at the University of Bergen, and forms a component of Dr Ella Walsh’s postdoctoral research. The project contributes towards the understanding of archaeological site formation processes at Blombos Cave, South Africa, and it fits within SapienCE research on the chronology and geoarchaeology of Blombos deposits.
Blombos Cave hosts an archaeological sequence that is well known for its evidence of modern human behaviour during the Middle Stone Age (100-70 ka). This sequence comprises sediment that is likely derived from a combination of the disintegration of bedrock and input from windblown sand, but the provenance has not been systematically tested. These different transport processes have implications for how cave deposits are dated, and so understanding the relative influence of sedimentary sources is important for establishing the timing of human occupation of Blombos cave. This research project will address this by analysing the mineralogy and grain size distribution of sediments within cave. We aim to establish how the contribution of sources has changed over time. The data produced in this research project will be useful for other SapienCE researchers investigating the degradation of rock art and preservation of biomarkers within the cave. Insight into site formation processes at Blombos will also have implications for how deposits are dated and interpreted in other cave sites along the coastline.
Aim/Objective
The objective of this research is to establish the sedimentary sources of Blombos Cave archaeological deposits and to reconstruct how the contribution of these sources has varied over the period of site formation from 100 ka to 70 ka. We will analyse the mineralogy and grain size distribution of a suite of samples from a sedimentary profile in the archaeological sequence. This data will be interrogated in the context of a corresponding analysis of samples from sedimentary sources: the Bredasdorp group calcarenites that comprise the cave bedrock, and sedimentary units within dune and palaeosol sequences across the coastline. The findings from this research will have implications for establishing the chronology of Blombos Cave sequence, for dating archaeological sequences in similar settings along the South African coast, and will provide important contextual information to other SapienCE researchers working at Blombos.
Understanding the source of sediment has implications for how we establish a chronology of human occupation of the cave during the Middle Stone Age. Sedimentary units of the archaeological deposits in Blombos Cave have been dated using optically stimulated luminescence dating (OSL) (Jacobs et al., 2006, 2020). This is a well-established method for reconstructing the timing of geomorphological events, and a key assumption of this method is that the material being dated has been sufficiently exposed to sunlight before being buried (Murray et al., 2021). This exposure would occur if the deposits are composed of wind-blown sediment from outside the cave, yet micromorphological work on Blombos sediments identifies abundant glauconite and indicates input from cave bedrock disintegration (Haaland et al., 2020). Breakdown of the bedrock contributes sand grains to the archaeological deposits that have not been exposed to sunlight, and if the contribution of this source is high, it could have implications for the chronology established (Roberts et al., 1998). Luminescence dating at Blombos has assumed that the dateable fraction of the sediment is derived from wind-blown sediment into the cave (Jacobs et al., 2006), but this has not been tested systematically. This research will also have implications for how other cave sites across the South African coastline are dated, where similar assumptions are made at Klipdrift Shelter (Henshilwood et al., 2004), Pinnacle Point (Jacobs, 2010), and Klassies River (Feathers, 2002).
These findings on site formation processes will provide data to other researchers working at Blombos, particularly for considering rates of erosion and weathering of the bedrock, and preservation of artefacts and biomarkers within the sediments. Blombos cave is known for its evidence of the emergence of symbolic culture (Henshilwood et al., 2011, 2018; D’Errico et al., 2023). Whilst rock art can be a key marker of symbolic culture, rock art dating to the Middle Stone Age has not been found at Blombos. The TaphArt project aims to explore the potential existence of rock art here through assessment of the potential for destruction by wind erosion. Our data on the mineralogical composition of sedimentary sources and quantification of the contribution of granular disintegration of bedrock to the archaeological deposits will help reconstruct the rate of cave wall erosion and the potential rate of removal of rock art by natural processes.
Within the archaeological sequence, biomarkers such as leaf waxes provide important information on vegetation assemblages and foraging practices during the MSA, but these biomarkers are variably preserved throughout the Blombos archaeological sequence (Gokturk et al., 2023). Our data will help test whether the variable concentration of leaf waxes is caused by a change in the relative input of coastal derived aeolian sediments (containing leaf waxes) and bedrock derived contribution (void of leaf waxes).

Methodology
This research project will target a vertical sequence of samples from the Blombos archaeological deposit (Fig. 1) to capture variability in sediment composition and character throughout time. We will use mineralogy and grain size distributions as a means of identifying populations of grains sourced from different processes
Mineralogy
The mineralogy of archaeological deposits will be analysed using X-ray Diffraction (XRD). This analysis will be undertaken at the University of Bergen using a Bruker D8 ADVANCE ECO XRD scanner. XRD is a method that identifies the phase of crystalline minerals through irradiation with X-rays (Pryor and Hester, 1969). Small aliquots of each sample will be prepared by grinding the loose sediment to a powder. Both the bulk and the clay fraction will be analysed using this method. We aim to interrogate variations in abundance of distinctive minerals, such as glauconite, to detail changes in source. Glauconite mineral grains within the archaeological sediments are suggested to be derived from granular disintegration of the cave walls (Haaland et al., 2020) and quantification of glauconite could provide a tracer of the contributions of bedrock and aeolian input. Mineralogy on aeolian-palaeosol sequences near the cave details variations in quartz, feldspar and carbonate content (Jacobs et al., 2020) and investigation of a suite of mineral types will be informative.
Grain size
We will then use the Mastersizer 3000 laser diffraction particle size analyser at the University of Bergen. This method will establish the grain size distribution of sediment samples suspended in water. Particle sizes ranging from nanometers to millimeters in size can be quantified through the detection of scattered light within the machine. This is a widely used approach in geomorphological research to understand transport processes (Kondolf and Piegay, 2016). To interrogate potential sources, we will use cluster analysis and canonical correspondence analyses (CCA) on the grain size and mineralogy data.