**2.2. Soil sampling and preparation**

A degraded communal rangeland site with homogeneous soil type and grazed extensively by livestock, a common and widespread land use practice in SSA [22], in the uplands of the Drakensburg region of South Africa (**Figure 1**) was selected because it exhibited a degradation

**Figure 1.** Location of the study area in the Drakensburg uplands of South Africa. Shown are images of soil surface coverage by vegetation for the different land degradation intensities or grass aerial cover from 0 to 5% (heavily degraded), 25 to 50% (moderately degraded) and 75 to 100% (non-degraded).

gradient varying from heavily degraded grassland with visible bare soils in the north to nondegraded grassland in the south. Such a state of degradation is a common feature of many communal rangelands in this part of South Africa.

Three categories of grass aerial cover were identified from surface soils across a degradation gradient of the communal rangeland site. A direct assessment was conducted based on vegetation cover [23], specifically grass aerial cover, which is the area of the ground covered by the vertical projection of the aerial portion of plants [24], to determine whether the land was degraded or not. Aerial cover was assessed by placing a 1 m×1 m plot frame at fixed intervals along each corresponding aerial cover category, while aerial cover of the plants in the plot was recorded as an estimate of the percentage of the total area [25]. The following grass aerial cover categories were established: 75–100% (Cov100), corresponding to non-degraded grassland; 25– 50% (Cov50), corresponding to moderately degraded grassland; and 0–5% (Cov5), corresponding to heavily degraded grassland. At each grass cover category, three sampling points were randomly selected, resulting in nine equidistant sampling locations along the degradation gradient. Four replicate soil samples 1 m apart at each sampling point were collected in a radial basis sampling strategy from a 0.05 m surface layer, giving a total of 12 samples per grass cover category and 36 soil samples along the degradation gradient. The surface layer was intensively sampled because the effects of land degradation on the quality of soil have been shown to be more pronounced in this soil layer [9, 26, 27]. In addition to these samples, triplicate core samples were also collected for bulk density using a 0.075 m diameter metallic cylindrical core (height=0.05 m) following a similar sampling strategy. Soil samples for bulk density were taken directly from the field to the laboratory and immediately oven-dried at 105°C to determine the oven-dry weight using the gravimetric method [28].

Once in the laboratory, field moist samples for soil aggregate stability were passed through an 8 mm sieve by gently breaking the soil along planes of weakness, air-dried and stored at room temperature before soil analyses. The remaining air-dried soils were ground to pass through a 2 mm sieve for further soil physical and chemical analyses.
