January 2016 Issue of Wines & Vines

Breeding for Drought-Tolerant Vines

Understanding and exploiting differences in root architecture

by Kevin Fort and Andrew Walker
Drought-tolerant roots
A Riparia Gloire vine displays lateral root growth following four weeks of green-house culture in a rhizotron. The green stem at top-center is a vine trunk.

California is a thirsty state with an erratic supply of water. The most populous state in the nation, California serves the water needs of approximately 38 million individuals--nearly 12% of the entire U.S. population. California added roughly 3 million people in the past 10 years, and estimates of growth for the next 10 years look similar. Taking into account that California's irrigation-fed agricultural sector is vital to the food needs of the state and nation, and considering that climate change could further destabilize an already unpredictable water supply, it is important to find ways to conserve water wherever possible.

practical winery vineyard

The expense of irrigating a vineyard impacts the grower's bottom line, and in dry years the concern can intensify from a question of cost to a question of supply. One example is a directive by the State Water Resources Control Board in 2009 for a 25% reduction in water use in Sonoma County and a 50% reduction in Mendocino County.11 Although these reductions were not solely directed at viticulture, an effort to reduce water use by grapegrowers was initiated using a public demonstration of the water savings possible through improved irrigation regimes and soil-moisture monitoring.6


    1. Arkley, R.J. 1963 “Relationships between plant growth and transpiration.” Hilgardia 34: 559–584.
    2. Bowen, P., C. Bogdanoff and B. Estergaard. 2012 “Effects of converting from sprinkler to drip irrigation on water conservation and the performance of Merlot grown on a loamy sand.” Am. J. Enol. Vitic. 63: 385–393.
    3. Blum, A. 2011 Plant breeding for water-limited environments. Springer, New York.
    4. de Wit, C.T. 1958 “Transpiration and crop yields.” Versl Landbouwk Onderz No. 64.6. Institute for biological and Chemical Research on Field Crops and Herbage: Wageningen, The Netherlands.
    5. Ezzahouani, A. and L.E. Williams. 1995 “The influence of rootstock on leaf water potential, yield, and berry composition of Ruby Seedless grapevines.” Am. J. Enol. Vitic. 46: 559–563.
    6. Greenspan, M. 2009 “Sonoma County Water Agency: vineyard water conservation demonstration project.” Web. Retrieved Aug. 20, 2015, from scwa.ca.gov/files/docs/conservation/Final%202009%20Vineyard%20Conservation%20Report.pdf.
    7. Grimes, D.W. and L.E. Williams.1990 “Irrigation effects on plant water relations and productivity of Thompson Seedless grapevines.” Crop Sci. 30: 255–260.
    8. Matthews, M.A. and M.M. Anderson. 1989 “Reproductive development in grape (Vitis vinifera L.): responses to seasonal water deficits.” Am. J. Enol. Vitic. 40: 52–60.
    9. McCarthy, M.G., R.M. Cirami and D.G. Furkaliev.1997 “Rootstock response of Shiraz (Vitis vinifera) grapevines to dry and drip-irrigated conditions.” Aust. J. Grape Wine Res. 3: 95–98.
    10. Nuzzo, V. and M.A. Matthews. 2006 “Response of fruit growth and ripening to crop level in dry-farmed Cabernet Sauvignon on four rootstocks.” Am. J. Enol. Vitic. 57: 314–324.
    11. Sonoma County Water Agency. 2009 State water resources control board order WR 2009-0027-DWR Term 15 status report: Milestone 1, July 15, 2009. Web. Retrieved Aug. 20, 2015, from scwa.ca.gov/files/docs/conservation/stateboard2009/071509-Milestone-1.pdf
    12. Stevens, R.M., J.M. Pech, M.R. Gibberd, R.R. Walker and P.R. Nicholas. 2010 “Reduced irrigation and rootstock effects on vegetative growth, yield and its components, and leaf physiological responses of Shiraz.” Aust. J. Grape Wine Res. 16: 413–425.
    13. Williams, L.E. 2010 “Interaction of rootstock and applied water amounts at various fractions of estimated evapotranspiration (ETc) on productivity of Cabernet Sauvignon.” Aust. J. Grape Wine Res. 16: 434–444.

The use of cultural practices to limit water use has been shown to have dramatic effects on the amount of crop produced per unit of water supplied. One study in the arid, sandy soils of the Okanagan Valley of British Columbia converted sprinkler-irrigated vineyards to drip and reduced applied water by approximately 64%.2 Although yields dropped approximately 40% in the first year, in the three subsequent years yields were equivalent. Presumably, in the first year of cultural change, the vines were adapting to the new conditions by increasing the root density in the small volume of soil wetted below the emitters.

But irrigation reductions usually come with a penalty. The strong relationship between water availability and crop productivity is, by now, an agricultural principle. A plant with more available water will transpire more during a growing season, and the relationship between transpired water and dry matter produced by plants per unit area can be described with simple formulas. This formula is simplest when conditions are relatively humid, wherein the relationship is: plant dry matter = total transpiration X n, where "n" is a constant value.1,4 This relationship is apparent in irrigation studies with grapes.7

When own-rooted Thompson Seedless was irrigated such that total evaporation plus transpiration was reduced approximately 25% from a maximum value, yield declined approximately 15%.7 When all irrigation was withheld, total evaporation plus transpiration was reduced by approximately 60%, and yield declined approximately 40%.7 In a Sonoma County field demonstration using soil-moisture probes and plant water-stress measures to better gauge water application, irrigation was reduced 60% to 100%, but yield loss occurred in all treatments, ranging from 13% to 42%.6

Complexities in water use

If pursued no further, the excellent studies that document the close relationship between transpiration and yield might lead to the premature and false conclusion that there really is nothing to be done: When water becomes scarce, yields will decline. Certainly this can be true. If nothing is changed other than applying less water, some loss in yield--even large losses in yield--should be expected. Fortunately, however, few issues in biology are ever simple, because simple trends are often components of a much larger, more complex system.

The first issue to address is water-use efficiency, or the yield obtained per total volume of water applied. The conversion to drip irrigation in virtually all California vineyards has been a major step forward in water conservation. As noted earlier, transitioning from sprinkler to drip irrigation increases water-use efficiency, sparing water that would otherwise be lost to weed growth, evaporation and soil wetting beyond the reach of the grapevine root system.2 Water-use efficiency was further improved in the above Sonoma County field demonstration by determining appropriate irrigation volumes with the use of soil moisture probes and plant stress measures to determine the timing of irrigation initiation and irrigation intervals.

The very meaning of drought is itself a complex issue. An equivalent seasonal precipitation could lead to water stress at one site with shallow soil, and therefore limited capacity for soil water storage, when compared to an adjacent site with a deeper soil profile. A moderately deep soil that is predominantly sand can have the same water storage limitations as a shallow loam. In reality, most sites are heterogeneous mixes of soil depth, texture, impenetrable layers, slope, aspect, stoniness and other factors, such that water stress can occur in patches even within the same block. The timing of drought is important. Flowering and set are particularly vulnerable to water stress in most crops, with early and late-season water deficits having relatively small impacts on final yield.

In a study using Cabernet Franc that imposed water deficits pre-véraison, post-véraison and throughout the growing season, post-véraison water deficit reduced yield a mere 2%, while pre-véraison deficit reduced yield 28%.8 The degree of water limitation is critically important, in that the modifications plants make to buffer yield losses when the yield potential is reduced by 5% to 40% are very different from the characteristics necessary to buffer yield potential reductions of 50% or more.3 Other factors muddying the picture are: the economics of pumping water when it is available, fruit quality benefits from imposing mild to moderate drought stress, longer term effects of soil salinization from repeated irrigations and expense of managing canopies when vine growth potential is high.

Breeding rootstocks for drought tolerance in grapes

Can a grapevine rootstock be bred that reduces or eliminates yield loss when water supplies are limited? If one approaches this question first through an examination of native vines that exist from the desert Southwest to the wet eastern United Sates, it is clear that multiple plant traits do exist that optimize plant growth across this large gradient of precipitation. Volumes have been written exploring these traits in detail, providing thorough explanations why, for example, a high-yielding sunflower would be doomed in the Mojave Desert without irrigation. The role of plant breeding for drought tolerance in grapevines then becomes a search for variability of these same traits within Vitis species, and combining multiple, favorable characteristics within the same plant to make a difference.

Because of the strong relationship between total transpiration and biomass produced, it is more important to produce a plant that is dehydration-resistant than to attempt to find a plant that can maintain productivity once drought has begun to desiccate plant tissues. Once leaf water potential begins to decline, stomata close more readily and carbon assimilation is reduced. At a given point, this effect is pronounced enough to impact yield. The studies cited, and many others, are evidence of this principle. However, if leaf--and plant--hydration can be maintained, or at a minimum if dehydration can be reduced, yield losses can be buffered or even eliminated.

There are two avenues to maintain leaf hydration: increased soil water access and the prevention of plant water loss. Effective soil water access is tied, at least in part, to root architecture and helps to explain how the leaf water potential of Ruby Seedless was minimally reduced when it was grafted on the deep-rooted 1103P rootstock.5 We are currently pursuing optimized methodologies for rapidly characterizing rootstock architecture. In a recent assay, 1103P had the steepest average root angle among all the commonly available rootstocks in California. Other deep-rooted rootstocks in this assay included Dog Ridge, 99R, Ramsey and St. George.

On the opposite end of the spectrum were rootstocks known to devigorate the scion, and these generally had shallow roots in our rapid assay: Riparia Gloire and 1616C. In addition to our rapid assay, we are characterizing root systems following excavation from greenhouse containers and from field plots. Although time-consuming both in the extended growth period and for data collection, this approach permits analysis of a more developed root system, allowing us to distinguish which defining architectural traits exist at the onset of root production and which require a greater degree of root system growth. How root system architecture changes in response to drought is a question that we are addressing through the excavations of vine roots.

In an ideal scenario, it would be desirable to observe root growth and architecture in real time, which is ordinarily hindered by opaque media and soil. But methods do exist that permit this on some level, and our method of choice is a rhizotron: a relatively two-dimensional container with a 2-foot-high by 1-foot-wide plexiglass face. Our rhizotrons permit us four to six weeks of observations before the vine has outgrown the container, a period long enough to observe growth patterns that correlate to the behavior of a genetically identical mature vine, and also long enough for us to observe root system responses to drought and recovery from drought. To date we have characterized nearly all of the commercially available rootstocks in California using this system and also completed a drought treatment using these rhizotrons.

It is possible to learn about the impact of a rootstock's roots without ever actually looking at the roots themselves. We grafted a common scion to seven rootstocks that span a range from very low to very high vigor potential. The resulting vines were planted in a test plot at the University of California, Davis, where a variety of irrigation methods will be used. Not only will we see the effect of the irrigation methods on such variables as yield, fruit quality and measures of plant stress, we will also observe the effect of various rootstocks. We have produced and are studying hybrid populations of rootstocks and various grape species with the goal of understanding the genetic basis of drought resistance and identifying more drought-resistant individuals from these populations.

Future directions

As is true in most scientific endeavors, much work remains to be done. Currently, effects of irrigation on yield far outweigh effects due to rootstock.11 What is unclear is why, in some cases, there are yield differences attributable to rootstock and in others there are not, and why evaluations that contain some rootstocks in common arrive at significant differences due to rootstock, but with different rootstock rank orders.5,9,10,12,13 When answers to these questions are found, rootstock breeding can more efficiently utilize the traits involved. Given that the majority of the world's rootstocks are derived from only three Vitis species, namely V. riparia, V. rupestris and V. berlandieri, the abundance of unexamined grape species, particularly those from the arid Southwest, holds great promise.

There is every reason to anticipate improvements in drought tolerance through continued efforts in grapevine rootstock breeding. First among these is that breeding for dehydration resistance has already been successful in other crops, especially rice and maize, often using straightforward and conventional methodologies. Grapevine breeding can follow suit. It is conceivable that advances made in other crops may be rapidly employed in grapes as genome sequencing becomes more widespread, rapid and less expensive. The drought tolerance catch phrase "more crop per drop" is already a reality when both improved cultural methods and appropriate rootstocks are employed. Further gains are largely a matter of continued effort to push the boundaries of what is known and explored, and each step forward will be needed as available resources diminish.

Kevin Fort is a postdoctoral research associate in the Department of Viticulture & Enology at the University of California, Davis. He studies salinity tolerance and drought resistance in grape rootstocks in the Walker lab and has studied salinity and drought in wildland plants from Mono Lake and Owens Valley.

Andrew Walker is a professor of viticulture at UC Davis as well as a grape breeder and geneticist. His work focuses on developing new rootstocks and disease-resistant wine grape varieties. Research for this project was funded by the California Grape Rootstock Improvement Commission; California Grapevine Rootstock Research Foundation; Fruit Tree, Nut Tree and Grapevine Improvement Advisory Board; California Table Grape Commission and American Vineyard Foundation. E. & J. Gallo Winery funded Dr. Kevin Fort's postdoctoral research.

Print this page   PRINTER-FRIENDLY VERSION   »
E-mail this article   E-MAIL THIS ARTICLE   »
Currently no comments posted for this article.