Determining the Best Cold Hardiness Measurement

Grapes contribute more than $5 billion per year to the economy of the eastern United States, and the grape and wine industries have been expanding rapidly in those regions. However, the sustainability of profitable yet cold-sensitive cultivars is limited by climatic constraints, primarily lethal freezing temperatures that can drop well below 0° F. The severe economic losses that result from freezing injury continue to be a major setback for the continuous growth of the grape and wine industries in this region.

In 2014, the polar vortex caused temperatures to drop to critical levels for grape survival and resulted in widespread damage and crop losses across the Midwest and eastern United States. Ohio experienced the worst damage and crop loss in 20 years (valued at $12 million). This problem was exacerbated with whole-vine loss of cold-sensitive species V. vinifera, which have dominated new plantings in the past 15 years in Ohio.

Therefore, the limitation imposed by the sensitivity of the vinifera species has an impact on the sustainable growth of the industry, and it is critical for growers to know the cold hardiness (CH) of the newly introduced varieties, in addition to the standard viticultural and enological characteristics. Sub-freezing temperatures below -20° C (-4° F), often damage V. vinifera, which is mostly adapted to a Mediterranean-type climate. Nonetheless, information about the CH of the newly introduced vinifera varieties has not been available.

Ohio’s cold-hardiness study The purpose of the study at the Ohio Agricultural Research and Development Center in Wooster was to determine the cold hardiness of 23 varieties of recently introduced V. vinifera grapes so that grapegrowers and nurseries would have a better understanding of the ability of those varieties to survive the harsh winter conditions in the Midwest and northeastern United States. Researchers also investigated whether there are better ways to estimate the CH of vines.

Buds were collected from each variety bi-weekly between September and March for three dormant seasons (2011-14) and placed in a freeze chamber. The freeze chamber was set up indoors to conduct freezing tests that would simulate cold damage in the vineyard. Cold hardiness was determined by thermal analysis and expressed by the lethal temperature that kills 50% of the bud population, or LT50.

We evaluated three methods to express cold hardiness: 1) Standard method: computing mid-winter hardiness based on a single and lowest LT50 value; 2) Modified standard method: computing mid-winter hardiness based on multiple lowest LT50 values; 3) Annual freezing tolerance (AFT, or annual cold hardiness): computing LT50 not only during mid-winter, but also during fall acclimation and spring deacclimation.

Cold hardiness of 23 varieties As expected, all 23 varieties behaved similarly and followed the typical U-shaped pattern of bud cold hardiness, which consists of three stages: fall cold acclimation (September to December), followed by maximum hardiness in mid-winter (January to February) and deacclimation in late winter and early spring.

However, there were differences of cold hardiness among varieties. Gamay Noir, Rotberger, Pinotage and Regent were generally among the most cold-hardy varieties with the lowest LT50 values, while Tempranillo, Lagrein, Barbera and Durif were among the most cold-sensitive varieties with the highest LT50 values.

Also, this study showed that not all varieties acclimated, de-acclimated or reached maximum hardiness at the same time and same level. “Cold Hardiness for Three Seasons” shows three varieties to illustrate this finding. Gamay Noir was the most cold-hardy variety (lowest LT50 in mid-winter) and was the quickest (steepest slope) to cold acclimate in the fall. Tempranillo was the most cold sensitive (highest LT50 in mid-winter) and was the quickest to deacclimate in late winter. Dolcetto acclimated the slowest in the fall and reached maximum cold hardiness later than Gamay and Tempranillo.

Methods of assessing cold hardiness Standard method: Mid-winter bud cold hardiness (LT50) is the most commonly used index to estimate a given variety’s cold hardiness and has been reported as such by research institutions in several states conducting this type of work in the past 30 years. In Ohio, we found that mid-winter cold hardiness varied between LT50 = -18.3° C (-0.9° F) and LT50 = -24.8° C (-12.6° F). “Cold Hardiness Based on Standard Method” shows that Gamay Noir was the most cold-hardy variety, whereas Tempranillo was the most cold sensitive in this study.

We also found that the standard method has a few drawbacks: a) the lowest LT50 in mid-winter is generally based on a single LT50 measurement; b) not all varieties reached their maximum cold hardiness on the same date; and c) wide variability of LT50 within each cultivar made it difficult to distinguish cold hardiness between two varieties if the difference between their LT50 was less than 3° C (5.4° F).

During the polar vortex in January and February 2014, the research vineyard in Wooster experienced successive low temperatures that dropped below the critical temperature of -20° C (-4° F) five times (-20.1° C [-4.2° F] to -24.5° C [-12.1° F]). All 23 varieties sustained trunk damage, with the exception of Regent. The extent of whole-vine damage was assessed in late summer 2014, and vine death was recorded.

In general, variety vine mortality corresponded to the computed LT50. For example, Tempranillo, with the highest LT50 (most cold sensitive), sustained the highest vine mortality (100%), while Gamay Noir, with the lowest LT50 (most cold hardy), sustained the least vine mortality (17%). Therefore, the 2014 winter damage sustained in the vineyard coincidently confirmed our cold hardiness determination in the laboratory.

Modified standard method: This method based on regression analysis was explored to alleviate the limitation of the standard method. It proved superior to the standard method since it distinguished the CH between two varieties at 2° C (3.6° F). However, it was not possible to obtain estimates for all varieties tested since more LT50 observations were needed.

A new index–the AFT: The annual freezing tolerance (AFT) or annual cold hardiness consisted of averaging LT50 throughout the dormant season. AFT has several advantages over the standard method currently used by nurseries, growers and researchers. First, AFT provides an estimate of CH across all stages of cold acclimation (i.e., autumn acclimation, mid-winter maximum hardiness and spring deacclimation). As a result, the AFT had the least LT50 variability, and differences between two varieties were detected at the 1° C (~ 2° F) level (see “Annual Freezing Tolerance for Calculating Cold Hardiness”). Second, the standard method provides an LT50 of a given variety determined on a specific date and location. Hence, this absolute value provides an arbitrary measure of CH, which will vary with time and space. In other words, one cannot extrapolate LT50 published values determined in one location (e.g., in New York) to equate the actual LT50 in another region (e.g., in Missouri). This may be common knowledge within the research community, but it is frequently misinterpreted by vineyard owners and nurseries.

To overcome this issue, we propose to compare the CH of varieties to a standard or benchmark variety grown in the same location. We call the new index the relative annual freezing tolerance, or RAFT. Merlotwas selected since it is a common variety grown in different regions worldwide and has mid-range CH.

The figure above illustrates RAFT with negative numbers indicating more cold-hardy varieties and positive numbers more cold-sensitive varieties than Merlot. For example, Gamay has a RAFT of -2.5° C, meaning Gamay is more hardy than Merlot by ~2.5° C. However, Barbera has a RAFT of ~1.5° C, meaning Barbera is less cold hardy than Merlot. In other words, if a region is known to have Merlot as a borderline variety for winter survival, then a grower should not even think about planting Barbera in the same region.

Finally, among the three methods, AFT produced the best prediction of whole-vine CH.