Previous PageTable Of ContentsNext Page

Regulated deficit irrigation and partial rootzone drying as irrigation management techniques for grapevines

M.G. McCarthy,
South Australian Research and Development Institute, Adelaide, Australia
B.R. Loveys, CSIRO Plant Industry, Adelaide, Australia
P.R. Dry and M. Stoll, University of Adelaide, Adelaide, Australia


Regulated deficit irrigation (RDI), an irrigation scheduling technique originally developed for pome and stone fruit orchards, has been adapted successfully for winegrape production. Water deficit is applied during the post-set period of berry development to reduce vegetative growth and, as necessary, berry size of red-winegrape varieties. However, water deficit is avoided during the berry-ripening period, and precise irrigation management is required to ensure minimal competition between ripening berries and vegetative growth. For the variety Shiraz, in particular, this irrigation practice has resulted in significant improvements in wine quality. Partial rootzone drying (PRD) is a new irrigation technique that improves the water use efficiency of winegrape production without significant crop reduction. The technique was developed on the basis of knowledge of the mechanisms controlling transpiration, and requires that approximately half of the root system be always in a dry or drying state while the remainder is irrigated. The wetted and dried sides of the root system are alternated on a 10- to 14-day cycle. PRD irrigation reduced significantly stomatal conductance of vines when compared with vines receiving water to the entire root system. Both systems require high management skills, and accurate monitoring of soil water content is recommended. Drip and other forms of micro-irrigation facilitate the application of RDI and PRD.

There is increasing global demand for high-quality wine and declining demand for wines of lower quality and lower value. Therfore, the challenge facing winegrowers is to improve winegrape quality.

In many regions, in particular in New World vineyards, irrigation is an integral feature of winegrape production. Traditionally, winegrowers have used irrigation maximizes productivity, as is reflected in recommended crop coefficients (FAO, 1977; FAO, 1998). Such coefficients help predict peak water requirement, and therefore are useful in the design stage of vineyard development. However, use of these values will result in water application rates in excess of those that may be optimal for the most appropriate balance between vegetative and reproductive development required for the production of premium quality grapes.

The key to improving winegrape quality in irrigated vineyards is to achieve an appropriate balance between vegetative and reproductive development, as an excess of shoot vigour may have undesirable consequences for fruit composition. Water stress has a major influence on shoot growth, and, in general, vegetative growth is more sensitive to water stress than is berry (fruit) growth. For some winegrape varieties, control of berry size is of importance. However, rrigation is not the only vineyard practice contributing to an inappropriate balance between vegetative and reproductive growth. Others include the use of rootstocks that impart high shoot vigour, improved plant nutrition and soil management, and the tendency to grow vines in cooler regions, which may favour vegetative growth at the expense of fruit growth. However, in many localities, the key to achieving the correct balance is irrigation management.

In recent years, the two main approaches for developing practical solutions to manipulate grapevine vegetative and reproductive growth have been: regulated deficit irrigation (RDI) and partial rootzone drying (PRD). However, these developments have been possible only as a consequence of better understanding of physiological responses to water deficit and the widespread use of drip and other forms of micro-irrigation that enable the precise control of water application rate and timing. RDI and PRD have become established water management techniques, both in New and in some Old World regions.

Regulated deficit irrigation

RDI uses water stress to control vegetative and reproductive growth. It was initially applied in peach and pear orchards to control growth by imposing water stress at key stages of fruit development. In an experiment on pear trees (Mitchell et al., 1989), irrigation application was reduced from 93 percent of the water evaporated from free water surface equivalent to the tree planting square to either 23 or 46 percent for a period of 19 d between November and December (southern hemisphere). After rapid fruit growth commenced, irrigation amount was returned to 120 percent. Compared with non-RDI trees, fruit growth was stimulated and vegetative growth reduced. The effectiveness of the RDI treatments was greater at higher tree density with the associated increased root competition.

In grapevines, reduced irrigation prior to veraison caused a greater reduction in berry size than did less irrigation after veraison, compared with control vines (Matthews et al., 1987). Wine made from fruit of continually drip-irrigated vines was unlike wine from early- or late-season deficit treatments, and distinctions were evident between `early-deficit' and `late-deficit' wines in appearance, flavour, taste and aroma (Matthews et al., 1990). Tasters of these wines indicated that `late-deficit' wines had a greater intensity of blackcurrant aroma compared with `fully irrigated' counterparts. The concentrations of anthocyanins and phenolics were higher in `deficit' wines although levels of residual sugar, titratable acid, pH and ethanol were similar to `fully irrigated' wines. The volume of water applied weekly to the least stressed treatment was about 50 percent of ETo for the site, and the most stressed vines received about 11 percent of ETo (Matthews and Anderson 1988).

In an experiment on winegrapes, Goodwin and Macrae (1990) reported that reduced irrigation during defined periods of berry growth after veraison reduced berry fresh and dry weights and sugar concentration. However, control vines in the experiment were not irrigated at full crop replacement and, consequently, in comparison to the initial work with RDI on stone and pome fruit, were deficit irrigated for the whole of the experiment. The question arises as to whether this practice is actually RDI. Less than full ETo replacement is often practised for the entire growing season in vineyards where water supply may be limited or in cooler districts where cropping level needs to be controlled to ensure adequate ripeness levels. In such situations, irrigation practice should perhaps more correctly, and simply, be termed `deficit irrigation.'

To compare the effects of water stress during berry developmental stages with well- irrigated vines, a large long-term field experiment was established on mature vines near Waikerie in the South Australian Riverland in the Murray-Darling Basin. Water stress treatments were imposed by withholding irrigation during four periods of berry development after flowering of Vitis vinifera cv. Shiraz. Control vines were irrigated such that water stress was minimized by regular monitoring of soil water content throughout the growing season and the use of a modern irrigation system that supplied water on demand (McCarthy 1997a, b, 1999, 2000). Coombe and McCarthy (2000) integrated these and previous findings into the figure reproduced here (Figure 1), with the addition of lines representing vegetative growth and the suggested period of water deficit. These findings were:

Figure 1
Idealized curves of vegetative and reproductive growth of Vitis vinifera

Note: dashed berry growth line represents weight loss period in Shiraz berries

Practical application

Many Australian winegrape vineyards normally use soil water monitoring to assist in the implementation of RDI. A variety of proven instrumentation is available. Where soil water content is measured with a neutron probe, for example, the available rootzone soil water content is kept below the irrigation refill line during the period of water deficit. The total available water in the rootzone should not decline by an amount greater than the difference between the full and refill lines (Figure 2). In practice, this may necessitate a light irrigation to prevent excessive water deficit. To control vegetative and reproductive growth, water stress should be limited to the period after fruit set in winegrape vineyards. This strategy is more applicable for red-wine varieties rather than white for which control of berry size and canopy size is considered less important. Monitoring shoot extension or comparing the rate of increase in berry weight with non-stressed vines can assess the effectiveness of the water deficit.

Figure 2
Use of a neutron probe to monitor changes in soil water content

In practice, reduced irrigation application during the post-set period may not achieve the desired outcome. The site may be unresponsive to irrigation due to factors such as:

There are various approaches for making sites with high water-holding capacity from winter rainfall responsive to post-set deficit:

The use of irrigation water containing moderate to high levels of salt (sodium chloride) may necessitate monitoring soil salinity during periods of reduced irrigation, and potentially the application of a leaching irrigation at the end of the period of reduced application. Other factors that may limit the successful adoption of water deficit during the post crop-set period are:

Conclusions - RDI

In Australia, numerous vineyards have adopted the concept of applying water stress immediately after fruit set to control vegetative growth, and, in particular for the variety Shiraz, to control berry size. In many instances this practice has resulted in significant improvement in red-wine quality, albeit sometimes at the expense of yield. In addition, experimental work has demonstrated that, contrary to the existing practice in many vineyards, controlled irrigation is recommended to avoid water stress during the fruit ripening period (post veraison). Minimizing water stress, whilst controlling vegetative growth, has resulted in more rapid ripening and a changed wine flavour profile. The continuance of controlled levels of irrigation during berry ripening is more necessary in drip-irrigated vineyards, where, as a result of drying of deeper soil layers and a reduced wetted soil volume compared with furrow irrigation, drought stress can rapidly develop during periods of high evaporative demand. This is particularly relevant in parts of Australia, United States of America, South America and South Africa, where the ripening period occurs under warm and dry conditions. Maintenance of higher levels of soil water content prior to, and after, harvest is now considered beneficial to post-harvest root growth and ensures vines do not enter dormancy under water stress, a condition that results in susceptibility to damage from cold weather. As a consequence, winegrape growers are now encouraged to use the term `strategic irrigation management' rather than RDI.

Partial rootzone drying

PRD uses biochemical responses of plants to water stress to achieve a balance between vegetative and reproductive development. By doing so, it achieves a secondary goal of significant improvement in production per unit of irrigation water applied. It has been a consistent feature of all trials that, even though the irrigation amount was halved, there was no significant reduction in yield due to PRD treatment. This contrasts with RDI experiments, where savings in irrigation application have often been at the expense of yield.

Research into the physiological changes that occur during water stress has led to improved understanding of plant response to stress in terms of chemical signals passing from roots to leaves. The vine's first line of defence when faced with water shortage is to close its stomata to conserve moisture. One of the principal compounds that elicits this response is abscisic acid. As soil water availability falls following the cessation of irrigation, this acid is synthesized in the drying roots and transported to the leaves in the transpiration stream (Loveys et al., 1999). Stomata respond by reducing aperture, thereby restricting water loss. Improvement in WUE results from partial stomatal closure. However, an inevitable consequence is reduced photosynthesis, as carbon dioxide and water vapour share the stomatal pathway through the leaf surface.

The challenge was to devise ways of controlling the amount of water available to grapevines, to maximize the production of root-derived chemical signals that reduce canopy transpiration, and, therefore, improve WUE. A methodology was developed to permit drying of part of the root system while keeping the remainder well watered. However, early attempts with grapevines were confounded by the transient nature of the response to drying part of the root system. By simply switching the wet and dry sectors of the rootzone on a regular basis, this transient response was overcome (Dry et al., 1996; Dry and Loveys, 1998).

A number of long-term, large-scale field experiments on Shiraz, Cabernet sauvignon, and Riesling, using a range of irrigation methods, have now been completed (Loveys et al., 1997, 1998, 1999; Dry et al., 2000). These included standard drip emitters (2 or 4 litres/h), two per vine in the inter-vine space and placed about 450 mm from the vine trunk (Figure 3) and subsurface drip lines, one on each side of the vine row at a depth of 200-250 mm and 350-400 mm from the centre of the row. In all cases, the intention was to create two wetted zones per vine that could be alternately irrigated on a cycle of approximately two weeks, i.e., while one zone was wetted, the other zone would be dried. Soil moisture sensors installed within each wetted zone assessed whether water applied to one side infiltrated to the other, supposedly dry, side. In all cases, there was satisfactory separation of wet and dry zones in a range of soil types under field conditions. Partial rootzone drying with furrow/flood irrigation has been successful in experiments with pears and citrus and in commercial vineyards in the Riverina district of New South Wales, Australia (Clancy, 1999), and with other perennial row-crop fruits.

Figure 3
Partial rootzone drying using two above-ground drip lines in a vineyard

In vines subject to PRD, there were reductions in vegetative growth as measured by pruning weight (Table 1). Much of the reduction in canopy biomass was due to reduced leaf area associated with lateral shoots. Total leaf area of PRD vines was significantly (P<0.05) less, largely the result of reduction in the area of leaves on lateral shoots (Figure 4). In another trial, minimally pruned Riesling vines were subjected to PRD and in July (southern hemisphere) the canes from three control panels (fully irrigated) and three PRD panels were removed and allocated to three length categories. Only the current season's growth was removed. The PRD treatment resulted in a significant reduction in the weight of canes in the >500-mm-size category and in the total pruning weight. Another measure of canopy density is the amount of light reaching the bunch zone and this figure was consistently higher in PRD than in control vines.

Table 1
Effect of PRD on pruning weight of vines, percent of control

Cabernet sauvignon - mean of four seasons


Minimally pruned Riesling (total all lengths)


   canes >500 mm


   canes 250-500 mm


   canes <250 mm


** Significantly different from control, P<0.01
ns Not significant

Figure 4
Main and lateral leaf area of control and PRD vines, post flowering and near harvest

A consistent feature of all trials was that there was no significant reduction in yield due to PRD treatment, even though irrigation amount was halved. As a result, yield per unit of water applied doubled in response to PRD (Table 2). Moreover, there was no effect on berry size in response to a halving of irrigation amount whereas there is usually a significant decrease in berry size in response to a substantial reduction in the amount of irrigation applied (Smart and Coombe, 1983; Williams and Matthews, 1990), particularly with deficit imposed between flowering and veraison (McCarthy, 1997a).

Table 2
Yield and water use of winegrapes





Shiraz, Adelaide, 1997-98

Yield (t/ha)
Water applied (Ml/ha)
Yield/Ml irrigation



Cabernet sauvignon, Adelaide, 1997-98

Yield (t/ha)
Water applied (Ml/ha)
Yield/Ml irrigation



Riesling, Waikerie, 1996-97

Yield (t/ha)
Water applied (Ml/ha)
Yield/Ml irrigation



Riesling, Waikerie, 1997-98

Yield (t/ha)
Water applied (Ml/ha)
Yield/Ml irrigation



The results of PRD on fruit composition in respect to wine-making attributes indicate that quality is at least maintained if not improved. Some experiments revealed no apparent effect on fruit quality as indicated by concentrations of anthocyanins and phenolics in fruit. In these cases, the control vines were well balanced with relatively open canopies: PRD did not substantially alter the canopy microclimate. In earlier experiments, the PRD treatment qualitative changes in the anthocyanin pigments of Cabernet sauvignon. For several seasons and at two sites, the concentration of the derivatives of delphinidin, cyanidin and petunidin in berries from PRD vines increased relatively more than the derivatives of malvidin and peonidin. Furthermore, PRD enhanced the formation of the coumarate forms of anthocyanins. This may be a response to bunch exposure, because shading of Shiraz bunches in a hot climate was found to enhance the proportion of coumarate forms.

Commercial trials have shown that if PRD is applied properly, there should be no significant yield reduction, although irrigation amount may be halved. A critical irrigation management practice with PRD is to ensure adequate rewetting of the dry side. Failure to ensure adequate replenishment of deep soil layers after switching sides may result in water stress, which may significantly reduce berry size during the early stages of berry development. Provided an overall favourable vine water status is maintained with PRD, berry size, and thus yield, will be maintained, despite reductions in water of up to 50 percent of conventional irrigation. A simple indication of whether the soil moisture status of the wet side was adequately maintained is the absence of reduction in berry weight. Similar to RDI, the responsiveness of the site to irrigation determines the successful application of PRD. Where the site is not responsive to irrigation, it is unlikely that part of the rootzone can be dried sufficiently during the initial stages of vegetative growth to control primary and lateral shoot extension. While savings in irrigation application may occur later in the season they may not be sufficient to economically justify the higher capital cost of installing PRD.

Conclusions - PRD

There has been much interest from New World viticultural industries in the PRD concept and its potential for influencing water use, vine vigour and grape quality. The implications for sustainable and profitable winegrape production are, well recognized. The successful adoption of PRD on a large scale has a number of consequences.

A reduction in consumption of water for irrigation is desirable from an economic viewpoint, although market forces will determine whether this ultimately translates to a reduction in district use or to the planting of additional vines or of other crops, to use the water saved. Further restrictions in water availability are probable and, in order to maintain productivity, irrigation practices and WUE will have to improve. Nevertheless, PRD does provide the vineyard manager with an additional management tool for tailoring crop quality to market needs.

The cost of implementing PRD varies depending on the irrigation system employed and whether it is applied to a new or existing vineyard. One of the most successful experiments in these projects utilized a pre-existing irrigation system consisting of two subsurface drip lines, one on each side of the row. In this case, the implementation cost was restricted to a few valves to allow switching water from one side to the other. At the other end of the cost scale, a development with the addition of a second drip line may cost about US$1 100/ha to install. Drip irrigation is in widespread use in vineyards throughout the world and, for example, in Australia a drip irrigation system may constitute half of the capital development cost. The additional outlay of installing PRD, is economical where the cost of irrigation water is high and as water becomes an increasingly valuable and scarce resource. The true environmental cost of irrigation water justifies the cost of implementing PRD.

The evaluation of PRD has progressed beyond the experimental stage with significant areas of PRD installed in vineyards in Australia, New Zealand, Spain, Israel, the United States, and South Africa. To date, most installations have involved a second drip line either above or below ground. Several irrigation-equipment manufacturers are working to eliminate the need to install two separate drip lines and to improve methods of installation and reduce root penetration in buried systems. Further research is underway in Australia to determine the optimum configuration for above- and below-ground installations, such as spacing of `on' and `off' drippers relative to vine spacing.

General conclusion

Table 3 summarizes the factors that determine the choice of RDI and/or PRD as an irrigation method in an individual vineyard.

Table 3
Relevant factors in choosing RDI or PRD as a vineyard management system



Site must be responsive to irrigation

Can be used with furrow irrigation

Drip irrigation preferred, alternate row furrow possible

Water must be available on demand

Control of berry size

No effect on berry size

Vegetative growth control

Vegetative growth control

Potential for yield loss

No loss of yield

Positive effects on grape and wine quality

Possible improvement in grape and wine quality

Marginal water savings

Significant water savings

No irrigation hardware modification

Significant changes required. Can be retrofitted.

Soil water monitoring recommended

High-level management skills required


Clancy, A. 1999. Riverina has the capacity to deliver diverse requirements. Australian Viticulture 3: 38-42.

Coombe, B.G. & McCarthy, M.G. 2000. Dynamics of grape berry growth and physiology of ripening. Australian Journal of Grape and Wine Research 6: 131-135.

Dry, P.R., Loveys, B.R., During, H. and Botting, D.G. 1996. Effects of partial root-zone drying on grapevine vigour, yield composition of fruit and use of water. In: C.S. Stockley, A.N. Sas, R.S. Johnstone & T.H. Lee, eds. Proceedings 9th Australian Wine Industry Technical Conference. Adelaide, Australia,Winetitles.

Dry, P.R. and Loveys, B.R. 1998. Factors influencing grapevine vigour and the potential for control with partial rootzone drying. Australian Journal of Grape and Wine Research 4: 140-148.

Dry, P.R., Loveys, B.R., Stoll, M., Stewart, D. & McCarthy, M.G. 2000. Partial rootzone drying - an update. Australian Grapegrower and Winemaker 438a: 35-39.

FAO. 1998. Crop evapotranspiration by R.G. Allen, L.S. Pereira, D. Raes & M. Smith. Irrigation and Drainage Paper No. 56, Rome.

FAO. 1977. Guidelines for predicting crop water requirements by J. Doorenbos and W.O. Pruitt. Irrigation and Drainage Paper No. 24. Rome.

Goodwin, I. & Macrae, I. 1990. Regulated deficit irrigation of Cabernet Sauvignon grapevines. Australian and New Zealand Wine Industry Journal 5: 131-133.

Loveys, B.R., Dry, P.R., & McCarthy, M.G. 1999. Using plant physiology to improve the water use efficiency of horticultural crops. Acta Horticulturae 537: 187-199.

Loveys, B.R., Grant, W.J.R., Dry, P.R. & McCarthy, M.G. 1997. Progress in the development of partial root-zone drying. The Australian Grapegrower and Winemaker 403: 18-20.

Loveys, B.R., Stoll, M., Dry, P.R. & McCarthy, M.G. 1998. Partial rootzone drying stimulates stress responses in grapevine to improve water use efficiency while maintaining crop yield and quality. The Australian Grapegrower and Winemaker 414a: 108-113.

Matthews, M.A. & Anderson, M.M. 1988. Fruit ripening in Vitis vinifera L.: Responses to seasonal water deficits. American Journal of Enology and Viticulture 39: 313-320.

Matthews, M.A., Anderson, M.M. & Schultz, H.R. 1987. Phenological and growth responses to early and late season water deficits in Cabernet franc. Vitis 26: 147-160.

Matthews, M.A., Ishii, R., Anderson, M.M. & O'Mahony, M. 1990. Dependence of wine sensory attributes on vine water status. Journal of the Science of Food and Agriculture 51: 321-335.

McCarthy, M.G. 2000. Developmental variation in sensitivity of Vitis vinifera L. (Shiraz) berries to soil water deficit. Australian Journal of Grape and Wine Research 6: 136-140.

McCarthy, M.G. 1999. Weight loss from ripening berries of Shiraz grapes (Vitis vinifera L. cv. Shiraz). Australian Journal of Grape and Wine Research 5: 10-16.

McCarthy, M.G. 1997a. Effect of Timing of Water Deficit on Fruit Development and Composition of Vitis vinifera cv. Shiraz (Ph.D. Thesis). Adelaide, Australia, Department of Viticulture, Horticulture and Oenology University of Adelaide.

McCarthy, M.G. 1997b. The effect of transient water deficit on berry development of cv. Shiraz (Vitis vinifera L.). Australian Journal of Grape and Wine Research 3: 102-108.

Mitchell, P.D., van den Ende, B., Jerie, P.H. & Chalmers, D.J. 1989. Responses of `Barlett' pear to withholding irrigation, regulated deficit irrigation, and tree spacing. Journal of the American Society for Horticultural Science 114: 15-19.

Smart, R.E. & Coombe, B.G. 1983. Water relations in grapevines. In: T.T. Kozlowski ed. Water deficits and plant growth. New York, United States of America, Academic Press Inc.

Williams, L.E. & Matthews, M.A. 1990. Grapevine. In: B.A Stewart and D.R. Nielsen, eds. Irrigation of Agricultural Crops. Madison, Wisconsin, United States of America, American Society of Agronomy.

Previous PageTop Of PageNext Page