South Africa Wine Technical Yearbook 2025

vineyard soils. 10 The initial P content in the Rawsonville sand was substantially higher than 114-153 mg/kg reported for the vineyard where the bulk soil was collected. 7 This suggested that the bulk soil was probably collected where a fertiliser spill had occurred. The initial P levels in the Rawsonville sand were substantially higher than the maximum of 20 mg/kg recommended Irrigation with municipal water had almost no effect on the P content in any of the soils (data not shown). The P content in the 10-20 cm layer of the Rawsonville sand tended to be higher compared to the top layer following the third diluted WWW irrigation, thereby indicating that attenuation of P did not occur in the top layer (Figure 1A). In contrast, irrigation with diluted WWW increased soil P substantially more in the 0-10 cm layer compared to the 10-20 cm layer of the Lutzville sand and Stellenbosch granite soil over the four simulated seasons (Figures 1B & 1D). This trend indicated that P attenuation occurred in the top layer of these soils. Although the P content in the 10-20 cm layer of the Stellenbosch shale tended to be lower after the first irrigation, it inclined at the same rate over time as in the 0-10 cm layer (Figure 1C). This indicated that no P attenuation occurred from the second irrigation onwards. The drastic decline of available P in the Rawsonville sand following the third WWW irrigation was probably due to fixation by Ca 2+ when the pH (KCl) exceeded seven (Figure 1A). The P could also have formed soluble complexes with the constituents in the WWW that were either leached out or susceptible to breakdown and P release by the Bray II reagent. 22 However, since no leaching occurred when irrigation was applied, 18 it could not have contributed to the decline in available P. In contrast to the Rawsonville sand, available P in the Lutzville sand increased as the pH (KCl) increased well above seven, where the diluted WWW was applied (Figure for grapevines growing in soils containing less than 6% clay. 10

soil P of four soils varying in parent material and clay content. Materials and methods Experimental layout Details of the pot experiment were given previously. 20 In brief, a pot experiment was carried out under a 20 m x 40 m translucent fibreglass rain shelter at ARC Infruitec-Nietvoorbij. Four different soils, namely a Rawsonville sand, Lutzville sand, Stellenbosch shale and Stellenbosch granite, were included in the pot experiment. The control treatment soils were irrigated with municipal water. For the wastewater treatments, WWW was diluted to a chemical oxygen demand (COD) level of 3 000 mg/L. Analyses determined in the irrigation waters. The P and pH (KCl were determined by a commercial laboratory according to methods described previously. Results During the period of the pot experiment, the P levels in WWW obtained from the same winery near Rawsonville and also diluted to 3 000 mg/L COD for a field experiment, were 4.8±1.6 mg/L. 7, 21 The amount of P applied when the grapevines growing in the sandy soil of the field experiment were irrigated with the diluted WWW was 1.3±0.4 kg P per hectare per irrigation. Based on these results, the annual application would amount to 9.4±2.6 kg P per hectare per year if six diluted WWW irrigations were applied. This indicated that the annual amount of P applied was relatively small and that it varied between seasons. The initial P contents were 217 mg/kg, 6 mg/kg, 8 mg/kg and 15 mg/kg, respectively, in the Rawsonville sand, Lutzville sand, Stellenbosch shale and Stellenbosch granite soils. With the exception of the Rawsonville sand, P contents in the four soils were in line with values expected for Water samples were collected prior to each irrigation. Due to a misunderstanding, P was not

1B). This trend suggested that the increasing amounts of sodium applied via the WWW increased the soluble PO 3- . In the case of the initially acidic Stellenbosch shale and granite soils (Figures 1C and 1D), the amorphous Fe 3+ and Al 3+ phosphates became more soluble as the pH (KCl) increased towards the optimum, as proposed previously. 11 Since P was not determined in the irrigation water, models to estimate the effect of irrigation with diluted WWW on soil P based on the amounts applied could not be created. However, the general variation in available P for the four soils could be illustrated with a plot of relative P, as calculated for each soil and layer, against pH (KCl) (Figure 2). After the fourth season, available P in the Rawsonville sand was still well above the norm of 20 mg/kg proposed for grapevines 10 in sandy soils (Figure 1A). However, this must be regarded as an atypical situation due to the initially high levels. In the more realistic scenario, P in the Lutzville sand only narrowly exceeded 20 mg/kg after the fourth simulated season (Figure 1B). After the fourth season, P in the Stellenbosch shale soil (Figure 1C) was well below the norm of 30 mg/kg for grapevines in soils containing more than 15% clay. 10 Likewise, P in the Stellenbosch granite soil (Figure 1D) was less than the lower threshold of 25 mg/kg for soils containing 6-15% clay. 10 The slow increase in P was probably due to the relatively low amounts of P applied via the diluted WWW. 7 Although the minimum thresholds were not reached, this does not rule out the possibility that they could be achieved if diluted WWW is applied over a longer period. However, if grapevines and cover crops absorb the P applied via WWW, the minimum thresholds might not be exceeded to the extent that no fertilisers will be required. Acknowledgements • This article is an output of WRC Project K5/1881, entitled “The impact of wastewater irrigation by wineries on soils, crop growth and

39

TECHNICAL YEARBOOK 2025