The character, origin and palaeoenvironmental significance of the Wonderkrater spring mound, South Africa

https://doi.org/10.1016/j.jafrearsci.2010.02.004Get rights and content

Abstract

Wonderkrater is a spring mound consisting entirely of peat in excess of 8 m thick. It has yielded a pollen record extending back over 35,000 years, which has provided one of the very few proxy climatic records for the interior of southern Africa in the Late Pleistocene and Holocene. The current investigation of the morphology and sedimentology of the site has revealed that the peat mound formed due to artesian conditions at the spring, but that accumulation of the thick peat succession was made possible because of clastic sedimentation on the surrounding piedmont which in turn was brought about by aggradation on the adjacent Nyl River floodplain. The peat mound has remained elevated relative to the surrounding piedmont for most of the 35,000 year period. Aggradation of the mound was slower during the Late Pleistocene than the Holocene (0.06–0.1 m/1000 year and 0.2–0.38 m/1000 year, respectively). Controlled archaeological excavations yielded a diverse late Pleistocene fauna preserved in peat and sand in the mound. A 1 m thick, coarse sand horizon at the base of the peat deposit contained a rich Middle Stone Age (>30 k year) lithic assemblage. The MSA sand layer likely represents an arid phase, suggesting the site’s antiquity as a place of refuge for Quaternary animals and the people that hunted them.

Introduction

Spring mounds have been described from many parts of the world. In their most common form, the mound is constructed from calcium carbonate precipitated from the spring water, perhaps with a contribution from aeolian dust. This type of mound is very widespread in the Great Artesian Basin of eastern Australia (Boyd and Luly, 2005). Equally common are spring mounds constructed from siliceous material, which are typically associated with geothermal springs, often in volcanic provinces. Less common are spring mounds constructed from peat, although examples have been described from Tasmania (Macphail et al., 2001), eastern Australia (Boyd and Luly, 2005), East Africa (Ashley et al., 2002, Owen et al., 2004) and the USA (e.g. Glasser et al., 1996). Peat normally accumulates below water as this reduces the rate of organic matter decomposition, and since water forms a horizontal surface, very special conditions are necessary to form a peat mound above the general terrain level, perhaps the most important being artesian conditions. Ashley et al. (2002) proposed the name ‘artesian blister wetland’ for the east African peat spring mounds, because artesian pressure sustains a small pool of water in the top of those mounds.

Spring sites have yielded important archaeological and palaeoenvironmental information. Examples include recovery of the skull of an archaic Homo sapiens and associated fauna from the Florisbad spring in South Africa (Dreyer, 1938), and evidence enabling the construction of the Holocene climatic record of the southwestern USA from the Montezuma Well (Blinn et al., 1994). But perhaps the most famous of the world’s peat spring mounds is Wonderkrater, situated in the Limpopo Province of South Africa (24°25′50″S, 28°44′36″E) (Fig. 1). This site has yielded pollen samples which have provided an almost continuous record of the vegetation, and by inference, the climate of central southern Africa extending back over the past 35,000 years (Scott, 1982, Scott, 1989, Scott, 1999, Scott and Vogel, 1983, Scott and Thackeray, 1987, Scott et al., 2003). Although the palynology of the peat mound has received detailed attention, a comprehensive geomorphic study of the mound has never been undertaken. The present study was carried out to rectify this deficiency and also because it was evident that the Wonderkrater site could provide important insights into the geomorphological development of the region and especially of the adjacent Nyl River floodplain (Scholes and Walker, 1993, Tooth et al., 2002), which hosts the Nylsvley Nature Reserve, an important RAMSAR site.

Section snippets

Geological setting

The study area is underlain by granites, felsites and mafic rocks of the early Proterozoic Bushveld Complex. These are unconformably overlain by horizontally-bedded sandstones of the mid-Proterozoic Waterberg Group, and sandstones, mudstones and basalts of the Permo-Triassic Karoo Supergroup (Fig. 1; Wagner, 1927, Geological Survey, 1978). The Karoo strata generally underlie topographically lower ground, partly a consequence of large-scale regional faulting, but also owing to considerable

Climate

The mean annual rainfall in the area is about 630 mm, but is very variable and over the past 90 years has ranged from 250 mm to 1100 mm (Scholes and Walker, 1993). Annual potential evaporation is about 2400 mm, resulting in a net water deficit. About 60% of the rain falls in the austral summer as localized convective thunderstorms and the reminder as a result of widespread frontal systems.

Regional geomorphology

The region is characterized by two distinct terrains. To the east of Wonderkrater lies the Springbok Flats, a vast, featureless plain lying at an elevation of about 1000 m amsl, with a very poorly developed drainage network. This region is largely underlain by rocks of the Karoo Supergroup. To the north and west lie the Waterberg Mountains, which are underlain predominantly by the resistant sandstones of the Waterberg Group and rise to an elevation of about 1800 m amsl.

Tributaries arise in the

Vegetation

The vegetation of the Wonderkrater spring has been described by Scott (1982) as broadly falling within Combretum veld that typically occupies the plains of the region, which is dominated by broadleaved deciduous trees and C4 grasses. The alluvial plains surrounding the Wonderkrater spring tend to be dominated by microphyllous woodlands with Acacia tortilis and Acacia karoo as dominant species, and with Acacia mellifera as an important component. The spring itself is dominated by Phragmites

Methods

Topographic and geomorphic information on the study area surrounding the spring mound were obtained from the 1:10,000 orthophoto map series published by the SA Government Chief Directorate: Surveys and Mapping. Detailed topographic, surface soil, and vegetation maps were made of the spring mound and its immediate surroundings using a plane table and level. Vegetation was sampled in circular plots of 2 m radius where herbaceous plants were dominant, and 5 m radius where woody plants were dominant.

Local geomorphology

The topography of the area surrounding Wonderkrater is shown in Fig. 2. To the south is the Tobias Spruit, one of the larger tributaries of the Nyl River. To the west are two smaller, unnamed tributaries that disappear on floodouts before reaching the spring mound. During wet periods, their discharge disperses around the spring mainly as sheet flooding with limited flow in poorly-defined depressions. Observations made after heavy rainfall reveal that the sheet flooding around the spring site

Discussion

The Wonderkrater mound has developed on an artesian spring which taps slowly circulating deep groundwater. Discharge appears to have been higher in the recent past when the spring formed the focus of a recreational resort, but there is currently no surface run-off from the spring mound. The present survey was carried out in a period of average to slightly above average rainfall, so the decline in discharge is more likely due to over exploitation of the aquifer that supplies the spring.

Conclusions

The Wonderkrater site is firmly entrenched in the archaeological and palaeoclimatic literature of southern Africa because it has been accumulating peat and pollen virtually continuously for the at least the past 35,000 years, and therefore provides a unique window into the climate of the region over this time. This is a particularly important time too, because it straddles the Pleistocene–Holocene transition. The present study has provided new insights into the mechanisms that led to sustained

Acknowledgements

The authors thank Dr. Walter Ward, owner of the site, for permission to work on the property and for generously accommodating us during our field work. We also thank farm manager Zacharia (Sakkie) Kekana for general assistance on the site, and Matt Kitching and Rhod McRae-Samuel for technical assistance. Financial support was provided by the National Research Foundation (NRF), University Research Council, University of the Witwatersrand, Palaeontology Scientific Trust (PAST) and Cultural

References (32)

  • T.F. Dreyer

    The archaeology of the Florisbad deposits

    Argeologiese Navorsing van die Nasionale Museum, Bloemfontein

    (1938)
  • Du Plessis, C.P., 1991. Tectonics along the Thabazimbi–Murchison Lineament. Unpublished PhD Thesis, University of the...
  • A.L. Du Toit

    Crustal movement as a factor in the geographical evolution of South Africa

    South African Geographical Journal

    (1933)
  • L.M. Fernandez et al.

    Seismic hazard maps for southern Africa

    (1992)
  • P.H. Glasser et al.

    Palaeo-reversals in groundwater flow and peatland development at Lost River, Minnesota, USA

    The Holocene

    (1996)
  • Geological Survey, 1978. 1:250,000 Geological Map, 2428 Nylstroom, South African Geological Survey,...
  • Cited by (20)

    • Untangling geochronological complexity in organic spring deposits using multiple dating methods

      2018, Quaternary Geochronology
      Citation Excerpt :

      However, in many springs artesian pressure sustains a water-table “dome” higher than the surrounding terrain, allowing organic material to accumulate above ground at spring vents when evaporation rates are lower than spring discharge (Ponder, 1986; Boyd, 1990a; McCarthy et al., 2010). Whilst under artesian pressure, capillary creep and slow outward diffusion of groundwater throughout these deposits is sufficient to maintain moisture levels minimising oxidation and desiccation (McCarthy et al., 2010). In addition to palaeoecological indicators organic spring deposits can preserve a range of other palaeoenvironmental information such as stable isotope ratios, geochemical tracers and rates of dust deposition through time (e.g. Dobrowolski et al., 2005; Dobrowolski et al., 2012, 2016; McGowan et al., 2012; Mazurek et al., 2014).

    • A late Quaternary record of monsoon variability in the northwest Kimberley, Australia

      2017, Quaternary International
      Citation Excerpt :

      An organic mound will only form at a spring where outflow is higher than the rate of evaporation, thereby preventing desiccation and mineralisation and where discharge is sufficiently low to prevent mound erosion (Ponder, 1986). Most commonly mounds are derived from aeolian deposits, precipitated calcium carbonate, or silicates in volcanic provinces (McCarthy et al., 2010). However, mounds comprised of peat will also form where there is sufficient water in otherwise challenging environments (Boyd, 1990a).

    • Fluctuations of vegetation and climate over the last 75 000 years in the Savanna Biome, South Africa: Tswaing Crater and Wonderkrater pollen sequences reviewed

      2016, Quaternary Science Reviews
      Citation Excerpt :

      Secondly, extrapolation or interpolation from and between dated sections was made at both sites. Thirdly, at Wonderkrater, local correlations between lithologies of the lower part of Borehole 4 (Scott, 1982a) and a recently dated excavation pit (Backwell et al., 2014; McCarthy et al., 2010) were made. These include radiocarbon and optical dates (IRSL and OSL) of a marked sand body and a lower clay horizon in excavation in Area B that were obtained by Backwell et al. (2014).

    • Multiproxy record of late Quaternary climate change and Middle Stone Age human occupation at Wonderkrater, South Africa

      2014, Quaternary Science Reviews
      Citation Excerpt :

      The mound aggrades as a result of the accumulation of peat, a brownish-black soil composed mainly of partially decomposed, loosely compacted organic matter that lived and died on the mound, and exists in a relatively fresh state of preservation because of the anaerobic waterlogged conditions. Because the spring is artesian, the accumulation of peat has created a topographically elevated feature that has excluded clastic sediment which forms the surrounding terrain (McCarthy et al., 2010). Based on a subsurface sediment map of Wonderkrater (McCarthy et al., 2010), four sites were selected for excavation (Areas A–D) (Fig. 2).

    View all citing articles on Scopus
    1

    Present address: Department of Geography, University of Melbourne, Melbourne, Australia.

    View full text