The motivation for effective irrigation of wine grapes is to maintain optimal yield while maintaining wine quality.
|As a result of limited water resources, producers should also attempt to achieve the highest possible yield with the least water without sacrificing wine quality. To be able to follow the correct irrigation strategy, the water requirements of grapevines should be known. The water requirement may differ from region to region, however, as a result of differences in soil type, climate, cultivation practices as well as the specific end product (Myburgh, 1998). Although the cultivation of wine grapes is an established industry in the Lower Orange River region, the water consumption, in other words evapotranspiration (ET), of wine grapes in this area, with its singular soils and climate, has not yet been determined. Previously the determination of ET and crop coefficients of wine grapes was limited to the Western Cape (Van Zyl, 1984a; Van Zyl & Fourie, 1988; Myburgh et al., 1996). Furthermore the published crop coefficients are only applicable to the American Class-A evaporation pans. It is therefore not applicable to the reference evapotranspiration that is nowadays calculated from weather data. The effect of different irrigation strategies on wine quality has not previously been determined in the Lower Orange River region.|
The purpose of the project was to determine how different irrigation cycle lengths will influence the growth, yield, wine quality and ET of grapevines on alluvial soils in the Lower Orange River region.
Material and methods
The investigation was carried out using Colombar/99R in the Gariep vicinity, situated between Upington and Groblershoop in the Northern Cape. At the start of the field trial the grapevines were eight years old. The grapevines were planted in alluvial soil, representative of the Dundee form (Soil Classification Work Group, 1991). The soil was deep ploughed before establishment using a wheel tractor. The plant spacing was 3.3 m by 1.8 m. The root system was well developed in the 600 mm deep silt rich topsoil. There were only a few roots in the sandy subsoil. This situation is common in the alluvial soils of the Lower Orange River region. The grapevines were developed on a T-trellis (Zeeman, 1981) and pruned to short bearers.
Four treatments were applied by irrigating once a week, every 14 days, 21 days and 28 days, respectively. Each irrigation treatment was replicated five times in a randomised block design. On each experiment plot, the row of six experiment grapevines was bordered by two grapevines at each end, as well as on both sides by two rows to prevent overlapping of treatment effects. The irrigation treatments were applied over four seasons (1996/97 to 1999/2000) from budbreak at the beginning of September to post-harvest in February. The vineyard was not cultivated mechanically during the duration of the field trial. Summer and winter weeds were controlled chemically over the entire surface.
Although the grapevines were planted in alluvial soil, irrigation was applied using micro-sprinklers instead of the more common flood method so that the irrigation quantities could be applied and measured more accurately. The flow rate of the micro-sprinklers (Eintal) was 32 L/hour at 100 kPa work pressure. The sprinklers were mounted on stakes so that the most even horizonal distribution of water possible could be obtained.
Soil water content was measured weekly using a neutron moisture probe at 30 cm, 60 cm and 90 cm depths. The probe was calibrated against gravimetric soil water content for the specific soil at each depth. Irrigation volumes were monitored using water meters. Initially the irrigation volumes were calculated using crop coefficients that had been determined for Sultanina in alluvial soil near Upington in the early nineties (Myburgh, 2003a). The irrigation quantities were gradually adapted to the different treatments by measuring the soil water content before and after irrigations. The universal water balance equation was used to calculate ET on a weekly basis as described by Myburgh (2003b). The atmospheric conditions and rainfall were measured using an automated weather station that had been erected at the trial. The reference evapotranspiration (ETo) was calculated from weather data using a modified Penman-Monteith equation (Allen et al., 1998). Crop coefficients were calculated by dividing the total ET for a specific week by the corresponding ETo.
Vegetative growth was quantified by measuring cane mass at pruning. The leaf area index (LAI), in other words square metre of leaves per square metre of soil surface, was measured at various stages of the growing season. Yield was determined by weighing the total amount of grapes on each trial site and converting this to tons per hectare. The total number of bunches per site were counted to calculate the bunch mass from the total crop mass. The berry mass was determined at harvest. For this purpose five berries from 20 different bunches on each plot were removed by using a pair of scissors. Sugar and acid content as well as the pH of the must were also determined during the harvest according to the standard procedures of the cellar at Nietvoorbij. Grapes from all the replications of each treatment were transported to Nietvoorbij, where experimental wines were vinified on a small scale according to the procedure described by Myburgh (2006a). The fermentation bouquet, acidity and overall wine quality were judged sensorially on an eight point scale each year in August by a panel of experts.
Results and discussion
Vegetative growth: By the end of September approximately half the vegetative growth had already taken place (Fig. 1). The leaf canopy of all the treatments was almost fully developed by the end of October. At that stage grapevines from the wettest treatment already had more leaves than those of the driest treatment. The wettest treatment (T1) had stronger shoot growth compared to the three drier treatments (Table 1). This was to be expected, seeing that shoot growth of wine grape cultivars, including Colombar, usually reacted readily to the water status in the soil (Van Zyl, 1984b; Myburgh, 2003b). In general shoot growth was stronger compared to the 2.4 t/ha for Sultanina in alluvial soil where irrigation was applied using furrows and drippers (Myburgh, 2006b). Seeing that it is not always possible in practice to apply canopy management in vigorously growing grapevines on the fertile alluvial soil, alternative irrigation strategies may be considered. Research has shown that irrigation may be applied every 14 days in alternative rows to restrict shoot growth of Sultanina in the Lower Orange River valley without sacrificing yield (Myburgh, 2003a). If a strategy such as partial wetting of the root zone may be applied successfully (partial rootzone drying = PRD), it could be beneficial where vigorous shoot growth on the alluvial soils may be detrimental to wine quality.
Yield: Berry size was reduced to such an extent in conjunction with an increased irrigation cycle length that berries of T2, T3 and T4 were significantly smaller compared to that of T1 (Table 1). A similar reaction was obtained where Colombar in the Breede River valley was irrigated at different soil water depletion levels (Van Zyl, 1984a). In general berries of all the treatments were smaller, however, compared to the approximately 1.7 g/berry of the driest level reported for the Breede River valley study. As a result of the smaller berries the bunch mass of T4 was lower than that of T1. The larger berries and heavier bunches caused the yield of T1 to be significantly higher compared to all the other treatments (Table 1). During the 1998/99 season hail destroyed almost 50% of the crop. Even so the yield of the grapevines that had been irrigated weekly was still higher than those which received the least amount of irrigation.
Sugar, acid and pH of the must: The grapes are usually ready for harvesting during the first week of February. During the 1996/97 season the average yield was approximately 60 t/ha. The grapes from the wettest treatment could only be harvested approximately three weeks after the other treatments. During the other seasons the average yields were lower and the tempo of ripening of all four treatments was more comparable so that all the grapes could be harvested on the same day. These results indicate that a combination of high crop loads and wet soil conditions may hamper ripening even in a warm, arid climate. The wetter soil conditions of T1 had no effect on sugar content, but delayed the reduction in acid (Table 2). Previous results showed that the wetter soil conditions are not only able to delay the reduction in acid in Colombar grapes, but that these may also cause sugar to increase more slowly (Van Zyl, 1984b). Despite the warm, dry conditions the acid content of the must was generally fairly high. The respective irrigation treatments did not have an effect on the must pH (Table 2). The must pH of all the treatments was within acceptable levels. Similar results were obtained with Colombar in the Breede River valley (Van Zyl, 1984b).
Wine quality: As a result of incomplete fermentation occurring in all the replications of the four treatments, experimental wines could not be made during the 1999/2000 season. Seeing that all viticultural practices, as well as the irrigation treatments, were applied just like the first three seasons, there is no explanation for the fermentation problem. The experimental wines were generally not of exceptional quality in any of the seasons. This could have been the result of negative effects of the warm, arid climate on the flavour intensity (Marais et al., 1999). The respective irrigation treatments did not have a significant effect on any of the wine quality parameters over the three seasons (Fig. 2). Previous research also showed the wine quality of Colombar to be fairly insensitive to irrigation (Van Zyl, 1984a). In all three seasons there was an obvious trend towards a more prominent fermentation bouquet as well as better overall wine quality in the case of the 28-day irrigation cycle (Fig. 2). It is important to note that the wine quality tended to increase when berry size decreased as a result of drier soil conditions. A similar tendency regarding wine quality and berry size also occurred in the coastal region of the Western Cape in Chenin blanc (Marais et al., 2005) and Sauvignon blanc (Myburgh, 2006a). This suggested that smaller berries as a result of reduced irrigation will not necessarily increase the quality of white wine dramatically. The results also showed that drier soil was only conducive to improved wine quality under a given set of circumstances and that irrigation per se was definitely not an overriding factor with regard to wine quality.
Fig 2: The effect of the length of the irrigation cycle on wine quality of Colombar as measured between the 1996/1997 and 1998/1999 seasons at the Gariep in the Benede-Orange River Region. Columns followed by the same letter do not differ significantly. (p=0.05)
Evapotranspiration and crop coefficients: During the last season the water meters with which the irrigation quantities had been measured, were damaged to such an extent that the ET could not be determined. In general the ET and crop coefficients decreased together with an increase in irrigation cycle length (Table 3). The smaller leaf area of the drier treatments (Fig. 1) probably contributed to the lower ET compared to the 7-day cycle (T1) seeing that less water could be lost through transpiration. Drier soil conditions, which reduce the transpiration tempo of grapevines (Van Zyl, 1984a; Myburgh et al., 1996), could also have contributed to the lower ET of the grapevines that were less regularly irrigated. Furthermore, regular, high evaporation losses from the wet soil in the case of the weekly irrigation (T1) may have increased the ET compared to the longer irrigation cycles (Myburgh, 1998). Research in Australia also showed that the ET of Colombar in drier soil was lower than in wetter soil (Stevens & Harvey, 1996). The ET difference between 21- and 28-day irrigation cycles (T3 and T4) was so small that it may be disregarded, however, and may therefore be considered the same for all practical intents and purposes. Although an accurate calibration was obtained with regard to the gravimetric soil water content, the neutron moisture probe was probably not sensitive enough to quantify the differences between T3 and T4.
The total annual ET of T2 and T3/ T4 was respectively approximately 14% and 25% less than that of T1 (Table 3). The total ET of the respective treatments was also considerably less than the annual allocation of 15000 m3 water quota per hectare in this particular area. The ET of the two driest treatments was comparable to the approximately 7000 m3 required to irrigate Sultanina in alluvial soil near Upington using furrows (Myburgh, 2003a; Myburgh 2006b), but was more than the 6000 m3 required for the irrigation of Colombar under cooler conditions in the Breede River valley (Van Zyl, 1984a). It should be borne in mind, however, that irrigation is more efficiently applied using micro-sprinklers than full surface flood irrigation. If vineyards in the Lower Orange River region are flooded weekly from September to February, the ET will probably exceed the allocated quota. If one assumes the efficiency of the micro-sprinklers to be 80%, the relationship between the amount of grapes produced and the amount of water applied, is 3.4 kg/m3, 3.6 kg/m3 and 3.9 kg/m3 for T1, T2 and T3/ T4, respectively. The water use efficiency in terms of the unit mass grapes per unit volume water was therefore lower than the 5.2 kg/m3 when Sultanina in alluvial soil near Upington was irrigated using furrows or drip irrigation (Myburgh, 2006c). The aforegoing also illustrates the extent to which water savings are possible, while retaining yield, if vineyards are not irrigated over the full surface.
The ET decreased with an increase in the irrigation cycle length. Where irrigation was applied every 21 or 28 days, there was no difference in ET for all practical intents and purposes. Shoot growth, berry size and yield also decreased with an increase in irrigation cycle length. There was a tendency towards improved wine quality when irrigation was applied less regular. If improved wine quality is the primary goal, and producers are rewarded accordingly, it may be financially rewarding to irrigate less. If this is not the case, however, it may be more profitable to maintain higher yields due to wetter soil conditions. The use of an irrigation strategy such as PRD to restrict shoot growth, but not the yield, must be investigated through ongoing research. If it can be applied successfully, a strategy like this may be beneficial where vigorous shoot growth on the fertile alluvial soils may be detrimental to wine quality.
The Agricultural Research Council for infrastructure and other resources, Winetech for partial funding, Mr Johan Coetzee for the use of his vineyard and the Soil Science personnel at ARC Infruitec-Nietvoorbij for technical assistance.
For more information contact Philip Myburgh on: email@example.com.
Allen, R.G., Pereira, L.S., Raes, D. & Smith, M., 1998. Crop evapotranspiration: Guidelines for computing crop water requirements. Irrig. Drain. Pap. 56, FAO, Rome.
Marais, J., Hunter, J.J. & Haasbroek, P.D., 1999. Effect of microclimate, season and region on Sauvignon blanc grape composition and wine quality. S. Afr. J. Enol. Vitic. 20, 19 – 30.
Marais, J., Van Schalkwyk, D. & October, F., 2005. Effek van korrelgrootte, sonligblootstelling en rypheid op Chenin blanc wynkwaliteit. Wynboer, August 2005, 14 – 15.
Myburgh, P.A., Van Zyl, J.L. & Conradie, W.J., 1996. Effect of soil depth on growth and water consumption of young Vitis vinifera L. cv. Pinot noir. S. Afr. J. Enol. Vitic. 18, 53 – 62.
Myburgh, P.A., 1998. Water consumption of South African vineyards: A modelling approach based on the quantified combined effects of selected viticultural, soil and meteorological parameters. PhD thesis, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa.
Myburgh, P.A., 2003a. Possible flood irrigation technologies to reduce water use of Sultanina grapevines in a hot, arid climate. S. Afr. J. Plant Soil 20, 1 – 8.
Myburgh, P.A., 2003b. Responses of Vitis vinifera L. cv. Sultanina to level of soil water depletion under semi-arid conditions. S. Afr. J. Enol. Vitic. 24, 16 – 24.
Myburgh, P.A., 2006a. Juice and wine quality response of Vitis vinifera L. cvs. Sauvignon blanc and Chenin blanc to timing of irrigation during berry ripening in the Coastal region of South Africa. S. Afr. J. Enol. Vitic. 27, 1 – 7.
Myburgh, P.A., 2006b. ‘n Ondersoek na moontlike waterbesparing met ondergrondse besproeiing I – Besproeiinghoeveelhede, benattingpatrone en wortelverspreiding. Wynboer, February 2007, 12 – 16.
Myburgh, P.A., 2006c. ‘n Ondersoek na moontlike waterbesparing met ondergrondse besproeiing II – Plantwaterspanning, groei, produksie en gehalte. Wynboer, March 2007, 13 – 16.
Soil Classification Work Group, 1991. Soil classification – A taxonomic system for South Africa. Dept. of Agricultural Development, Private Bag X144, Pretoria 0001, South Africa.
Stevens, R.M. & Harvey, G., 1996. Soil water depletion under large grapevines. Aust. J. Grape Wine Res. 2, 155 – 162.
Van Zyl, J.L., 1984a. Interrelationships among soil water regime, irrigation and water stress in the grapevine (Vitis vinifera L.). PhD thesis, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa.
Van Zyl, J.L., 1984b. Response of Colombar grapevines to irrigation as regards quality aspects and growth. S. Afr. J. Enol. Vitic. 5, 19 – 28.
Van Zyl, J.L. & Fourie, A., 1988. Estimating the irrigation requirements of vineyards using crop facors and the Class A-Pan. Farming in South Africa, VORI 227/1988 Dept. of Agricultural Development, Private Bag X144, Pretoria 0001, South Africa.
Zeeman, A.S., 1981. Oplei. In: Burger, J.D. & Deist, J. (eds.). Wingerdbou in Suid-Afrika. Nietvoorbij, Stellenbosch, South Africa, pp. 185 – 201.
The effect of different irrigation cycles on growth, yield, wine quality and evapotranspiration (ET) of wine grapes was determined over four seasons in a field trial at Gariep in the Lower Orange River region. Four treatments were applied by irrigating Colombar/99R with micro-sprinklers at 7-day, 14-day, 21-day and 28-day intervals, respectively. Irrigation cycles longer than 7 days not only reduced vegetative growth, but also resulted in smaller berries and reduced yields. Although overall wine quality was moderate under the warm climatic conditions, it tended to improve with irrigation cycle length. This trend was consistent over seasons. In December, peak ET for the 7-day, 14-day and 21-day irrigation cycles amounted to 5.5 mm/day, 4.6 mm/day and 4.2 mm/day, respectively. There was no practical difference in ET between 21-day and 28-day irrigation cycles. Mean annual gross irrigation requirements amounted to 12775 m3, 11012 m3 and 9612 m3 per hectare for the 7-day, 14-day and 21-day irrigation cycles, respectively. These volumes constituted a saving compared to the annual allocation of 15000 m3 for the particular region. However, even for grapevines receiving the least water, mean yield still remained in excess of 36 t/ha over the four seasons. This suggested that considerable water saving may be made without detrimentally affecting yield and wine quality.