**2.4. Determination of chemical and physical properties**

Particle size distribution was determined by the sieve and pipette method [30]. Soil pH was determined in a 1:2.5 solution ratio in both deionized water and 1 M KCl suspension using a Calimatic M766 pH meter. The exchangeable cations Ca and Mg were determined by extraction in 1 M KCl, while P, K, Zn, Mn and Cu were determined by extraction in an Ambic 2 extract containing 0.25 M NH4HCO3 [31], with detection by inductively coupled plasma optical emission spectrometry (ICP-OES) using an Optima 7300DV spectrometer (Perkin Elmer, Inc., Shelton, CT). Effective cation exchange capacity (ECEC) was calculated as the sum of extractable cations, with base saturation calculated as the proportion (%) of the ECEC accounted for by exchangeable bases (Ca, Mg, K and Na).

#### **2.5. Determination of soil aggregate stability**

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

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

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

In the field, penetration resistance (PR) was evaluated by randomly selecting 15 positions in each grass aerial cover category, and PR readings were taken in the topsoil surface layer. The PR of the soil, which is a proxy for soil compaction, was determined using a handheld cone penetrometer [29]. Notably, PR measurements were taken before the soil surface was disturbed

Particle size distribution was determined by the sieve and pipette method [30]. Soil pH was determined in a 1:2.5 solution ratio in both deionized water and 1 M KCl suspension using a Calimatic M766 pH meter. The exchangeable cations Ca and Mg were determined by extraction in 1 M KCl, while P, K, Zn, Mn and Cu were determined by extraction in an Ambic 2 extract containing 0.25 M NH4HCO3 [31], with detection by inductively coupled plasma optical

communal rangelands in this part of South Africa.

84 Land Degradation and Desertification - a Global Crisis

oven-dry weight using the gravimetric method [28].

**2.3. Penetration resistance**

for soil sample collection.

a 2 mm sieve for further soil physical and chemical analyses.

**2.4. Determination of chemical and physical properties**

After field sampling, moist soil samples were taken to the laboratory and air-dried at room temperature. During this period, large soil aggregates were periodically broken down by hand along lines of weakness to obtain maximum millimeter-sized aggregates. Soil samples were then sieved to isolate 3–5 mm aggregates for aggregate stability testing. Soil aggregate stability was determined on the 3–5 mm aggregates following the ISO standard method (ISO/DIS 10930:2012) outlined by Le Bissonnais [32]. The aggregates were subjected to rapid wetting by immersion into water, slow wetting by capillarity and mechanical disaggregation by shaking after wetting with ethanol, which correspond to different aggregate breakdown mechanisms, *viz*. slaking, differential clay swelling and mechanical breakdown, respectively. For the rapid wetting test, 10 g of 3–5 mm aggregates was submerged in 50 ml of distilled water in a beaker for 10 minutes, resulting in slaking of the soil. For the slow wetting test, 10 g of 3–5 mm aggregates was spread on top of a foam soaked in water. Thereafter, aggregates were allowed to wet through capillarity for 60 minutes. For the mechanical disaggregation test, 10 g of 3–5 mm aggregates was first immersed in a beaker with ethanol and then transferred to a beaker with distilled water to rest for 30 minutes. The aggregates were then transferred to an Erlenmeyer flask using distilled water and gently shaken up and down by hand 10 times. The weights of the aggregates collected on each sieve size (2, 1, 0.5, 0.2, 0.1 and 0.05 mm) were measured and expressed as the percentage of the initial dry mass sample. The mean weight diameter (MWD) for each disaggregation mechanism was calculated using the following equation:

$$\text{MWD} = \frac{\sum (\mathbf{x}\_i \mathbf{w}\_i)}{100},\tag{1}$$

where *x* is the mean inter-sieve size and *wi* is the percentage of fragments retained by the sieve *i*. The greater the MWD, the more resistant the soil aggregates are to the aggregate breakdown mechanisms.

#### **2.6. Statistical analysis**

Results are presented as standard error (SE) of the means for each grass cover along the degradation gradient and, where specified, subjected to one-way analysis of variance using GenStat (VSN International, Hemel Hempstead, UK). Differences between means were tested using Duncan's multiple range test at *P*<0.05.
