The chemistry of the pedoderm – part 3: Colophospermum mopane shrublands and woodlands in the central Kruger National Park, South Africa

The woody plant, Colophospermum mopane, occurs as either a shrub or tree in shrublands and woodlands, respectively, in the central Kruger National Park, South Africa. As a first step in identifying which soil properties potentially influence the structure of C. mopane in these landscapes, we investigated the soil chemistry of the pedoderm in shrublands and woodlands. A total of 39 composite pedoderm samples (0–2 cm) were analysed. Relative to the woodlands, the pedoderm of the shrublands had significantly lower mean concentrations of P, K, Ca, Mg, S, Mn, Cu, Na, B and C. Of the nutrients analysed, only B had a mean concentration likely to be constraining the growth of plants (0.16 ± 0.01 vs 0.69 ± 0.05 mg B kg−1 in shrublands and woodlands, respectively). We suggest that the C. mopane shrubs do not establish as trees in the shrublands because deficiency of B limits their growth and increases the competitive strength of the grass sward relative to eudicot plants in the landscape.

This paper is one of four research notes in this journal volume in which we explore the differences in soil nutrient status between treed and treeless areas in South African grasslands and savannas (Mills et al. 2021a(Mills et al. , 2021bMills and Kellner 2021). Our rationale for focussing on soil nutrient deficiencies in the pedoderm is explained in the first research note of the series. Here we report on the results from a site in the central Kruger National Park, South Africa. The woody plant, Colophospermum mopane, known locally as mopane, occurs as a shrub less than one metre tall, as well as a tree up to 18 metres tall, in the central and northern regions of the park (Van Wyk 2008). In many shrublands and woodlands in these regions, C. mopane is abundant (Figure 1). Given that these vegetation types experience the same climate, it is likely that the marked differences in vegetation structure are ultimately because of differences in soil properties, with herbivory and fire potentially playing a modifying role.
As a first step towards identifying the soil properties underpinning the differences in structure of the C. mopane shrublands and woodlands, we investigated the chemistry of the pedoderm in both habitats. The shrublands and woodlands investigated were located on flat plains, with clayey topsoils derived from Timbavati gabbro and Letaba basalt parent materials, respectively (Council for Geoscience South Africa 2017). Based on global datasets, satellite imagery analysis and machine learning, soil types in the shrublands have been classified as Eutric Regosols and Ferralic Arenosols, and in the woodlands as Eutric Cambisols and Rhodic Nitisols (ISRIC 2008), with mean clay content of the two vegetation types ranging from 23% to 29% (Hengl et al. 2017) (Supplementary Figure S1). With respect to herbivory and fire, the data available indicates that there are no differences between the shrublands and woodlands in our study area (Supplementary Figures S2 and S3).

Research Note
The chemistry of the pedoderm -part 3: Colophospermum mopane shrublands and woodlands in the central Kruger National Park, South Africa 0.02 M di-ammonium ethylenediaminetetraacetic acid (Trierweiler and Lindsay 1969;Beyers and Coetzer 1971). The Walkley-Black method (Walkley 1935;Nelson and Sommers 1982) was used to determine organic carbon (C). Differences between the vegetation types were tested using independent t-tests in R (version 3.6.1).
The potential importance of B on the growth of trees in South African savannas has previously been highlighted by a nursery experiment where the indigenous tree Vachellia karroo (Hayne) was grown in ten different topsoils. Boron was found to be the nutrient most strongly correlated with growth (Wakeling et al. 2010). The results of this nursery experiment and the differences in chemistry of the pedoderm between mopane shrublands and woodlands at our study site suggest that further research on B is likely to be a fruitful avenue for investigating the effects of nutrient availability on vegetation structure in South African savannas.
The origin of the differences in soil chemistry between the shrublands and woodlands at our study site is potentially confounded by the effect of plants on soil nutrients. It is plausible, for example, that the greater stature of trees compared with shrubs caused the enrichment of nutrients in the woodland, given that trees are known to enrich soils beneath their canopies (Scholes and Archer 1997;Ludwig et al. 2001;Augusto et al. 2002;Reis et al. 2010;Ruwanza 2017). We suggest, however, that differences in geology between the shrublands and woodlands is a primary factor causing the difference in vegetation height, with soil texture and nutrient enrichment under tree canopies being secondary factors. Parent materials of the study site include gneiss and gabbro under the shrubland and basalt under the woodlands. Nutrient richer and more clayey soils would be expected to develop from the basalt parent material. The more clayey texture would result in reduced rates of nutrient leaching, further widening the difference in nutrient status between the basalt-derived and gneiss/ gabbro-derived soils. The presence of a woodland on the basalt-derived soils would also increase this difference in nutrient status because of inter alia the capture of dust in the tree canopy and importation of nutrients in dung and urine from animals using the tree canopy for shade. Notwithstanding the effects of clay and tree canopies, the extreme differences in mean concentration of nutrients such as Ca (821 ± 114 versus 6 255 ± 845 mg kg −1 in shrublands and woodlands, respectively) suggests that geology has had a primary role in determining the nutrient status of the soils in the study area. Slight differences in clay content and nutrient enrichment by tree canopies would not be expected to generate an eight-fold difference in nutrient status. We consequently conclude that variation in geology, resulting in a relative deficiency of B in the shrublands, is likely to be a primary underlying reason for the major differences in vegetation structure in the study site investigated.