BY Wayne F. Wilcox
Department of Plant Pathology, Cornell University,
NY State Agricultural Experiment Station, Geneva, NY

Botrytis bunch rot (BBR) occurs sporadically in Mediterranean climates such as California, yet it can still cause significant crop losses in a challenging year, as 2006 so rudely demonstrated. Botrytis is a chronic problem in New York, as it is in other humid growing regions of the world. Therefore, we have devoted considerable effort to better understanding its biology and control.

Some of this research has been supported by the California wine industry through the American Vineyard Foundation, in addition to the U.S. Department of Agriculture Viticulture Consortium-East. Such support is gratefully acknowledged.

Disease biology
BBR is an amazingly complex disease, and is a virtual “poster child” for the concept of the so-called “disease triangle,” the idea that plant diseases are governed by multiple three-way interactions among the host (grapevine), the environment, and the causal pathogen (the Botrytis fungus).

Many of these interactions are poorly understood. This is fascinating to a research plant pathologist, but frustrating to a vineyard manager, who needs to determine when intensive disease control is necessary and when it’s not. Such complexity notwithstanding, there are a several basic concepts that go a long way to help understand this disease, and devise a strategy to manage it.

Botrytis cinerea is very widely dispersed, and can overwinter as resting structures on infected canes, or more commonly, in various vineyard debris. Old cluster stems appear to be a particularly common source of inoculum. Botrytis is a “weak” pathogen that primarily attacks highly succulent, dead, or injured tissues, or tissues that are senescing.

Berries damaged by insects, powdery mildew, or splitting, due to pre-harvest expansion and/or rain, are common injury sites attacked by Botrytis. Withering blossom parts and ripening fruit are senescing tissues of particular importance to the initiation and development of BBR.

The fungus thrives in high humidity and still air, hence the well-known value of cultural practices such as leaf removal and canopy management to minimize these conditions within the fruit zone.

Although the fungus does not grow well inside berries until they start to ripen, it can gain entrance into young fruit through senescing blossom parts, old flower “trash” sticking to berries within the cluster, and scars left by the fallen caps.

Infection requires that tissues remain wet for at least several continuous hours (rain). The ideal temperature for infection is approximately 60–75ºF, although it can occur at temperatures outside this range under longer wetting periods.

Infections early in the season remain confined to a few cells and latent (dormant) while berries are green, but given the “right” conditions, some of them can resume activity, and progress to the point that they rot the affected berries as the grapes approach maturity. Once this occurs, the disease can spread rapidly from berry to berry within clusters and, to a lesser extent, among clusters.

The likelihood and magnitude of disease spread depends on weather conditions and certain physiological factors within the vine.

The following are some recent findings concerning various details of this basic scenario, and the fungicides that are used to control the disease. Although this work was conducted in the Finger Lakes region of New York, the general principles should be applicable to California as well, although certain specifics will obviously be different.

Latent infections can be common following a wet bloom period, but the vast majority of latent infections appear to remain inactive through harvest, meaning that the fruit stays healthy. For example, when we’ve inoculated Pinot Noir clusters with Botrytis spores at bloom, we’ve been able to detect latent infections in up to 70% of the young berries that we’ve sampled at various stages thereafter.

However, we often end up with only 2% or 3% rotten berries at harvest, especially in a dry autumn or a clone with loose clusters. On the other hand, we’ve ended up with 25% to 40% rot at harvest in clones with tight bunches, particularly in wetter harvest periods; leading to the following point:

Serious Botrytis losses result from disease spread during the pre-harvest period, after berries begin to ripen and become highly susceptible to rot by the fungus. Thus, latent infections established at bloom can play a critical role in disease development if even a few infections become active and provide an initial “foot hold” from which spread can occur under favorable pre-harvest conditions.

Sprays to control one, both, or neither of these two phases of disease development may be worthwhile, depending on the specific vineyard characteristics and weather conditions.

The potential importance of early disease establishment, and two vine factors that promote subsequent disease spread, are illustrated by the results of two following field experiments.

In the first experiment, we inoculated one, three, or five individual Pinot Noir berries per cluster, 10 days after veraison, by using a hypodermic needle to inject them with Botrytis spores. This produced initial “point sources” of the disease within clusters about a week later, which were meant to simulate occasional activations of latent infections before harvest.

To determine the effect of cluster architecture on disease spread from these sources, we utilized clone 29 (very tight bunches) and opened up some of the clusters by removing enough berries after fruit set so that they resembled the Mariafeld clone (loose clusters, generally less prone to bunch rot than other Pinot Noir clones under commercial conditions).

In Figure I, Botrytis was able to spread extensively throughout the unthinned clusters. One single rotten berry that appeared 21/2 weeks after veraison was all it took to end up with 50 additional rotten berries at harvest (pre-harvest weather was wet that year).
In contrast, disease spread was minimal in the thinned clusters. Additional experiments showed that the relative Botrytis resistance long noted for the Mariafeld clone can be accounted for simply by its loose bunches rather than some inherent chemical factor.

In the second experiment, clusters of a tight-bunched Chardonnay clone were similarly thinned and inoculated. Additionally, some vines received four weekly sprays of urea (8 lb urea in 100 gal of water per acre), starting at veraison, to see if high berry nitrogen content would affect disease spread.

This treatment increased the mean level of assimilable nitrogen in the must to 303 mg/L, compared to 235 mg/L in the untreated plots (determinations from 10 clusters per each of four replicate plots per treatment; differences between the two treatments were statistically significant at P = 0.05).

In Figure II, little disease spread occurred in the thinned clusters, regardless of nitrogen treatment. In contrast, elevated berry N did increase spread in intact clusters with either one or three initial points of infection. The late season nitrogen applications did not increase canopy growth. Therefore, it appears that their effect on disease development was due to an increased susceptibility of berries to fungal colonization as the nitrogen content increased, rather than an increase in canopy density and associated microclimate effects.

In Figure II, the analysis of variance showed a highly significant (P<0.001) interaction between the thinning (opening up) treatment and nitrogen treatment.

Factors that determine whether latent infections become active and cause disease, or remain latent and symptomless, are poorly understood. However, the activation of latent infections appears to be influenced by both the weather and vine physiology. Experience and controlled experiments both show that high atmospheric humidity during a several day period prior to harvest is one such factor that promotes this process.

To examine this factor, we inoculated clusters of potted Chardonnay vines at bloom in order to establish latent infections. The vines were kept outdoors, protected from rain.While all clusters were still symptomless, we moved them into a high humidity (92% RH) chamber for 0 to 9 days either at veraison, or starting 10 days before harvest.

In Figure III, the percentage of clusters with disease symptoms (due to activation of latent infections) increased in proportion to the duration of the high humidity treatment pre-harvest, but was not affected by high humidity at veraison.

In addition to high atmospheric humidity, high soil moisture can also promote activation of latent infections. The influence of this factor is illustrated by the following experiment, in which potted Chardonnay vines were once again inoculated at bloom to establish latent infections and grown in a screenhouse, where they were protected from rain and irrigated as needed to maintain moderate growth and avoid drought stress.

At veraison, the vines were split into two groups: a) watered several times per week to keep the soil moisture levels constantly high (wet); or b) watered only when shoot tips began to wilt (dry). On these occasions, water was provided to alleviate stress, but the soil was not saturated.

Pots receiving the two irrigation regimes were interspersed among each other randomly. Therefore, all clusters were subjected to the same atmospheric environment, and only the soil environment was different.

Disease incidence was determined in two manners: 1) at the time of harvest; and 2) after harvested clusters were incubated for four days at 68ºF and 92% RH, to promote additional activation of latent infections. There were seven replicate plants with six clusters per plant for each of the two treatments.

In Table I, disease incidence (the best indicator of latent infection activation), was more than three-fold greater at harvest in the wet compared to dry soil treatment. However, most of this difference was no longer evident after post-harvest incubation, by which time there was no statistical difference between the two treatments.

This “leveling off” between the two treatments was due to the pronounced increase in disease incidence for clusters from the dry, but not the wet, treatment following incubation and represents the activation of multiple latent infections during the post-harvest incubation period.

Thus, it appears that the wet-soil treatment provided conditions that were more conducive to the activation of latent infections than did the dry treatment while clusters were on the vine, but that many latent infections in the dry treatment remained viable and activated upon exposure to high atmospheric humidity during incubation.

As for most fungi, Botrytis spores require a film of water in which to germinate and initiate infections. Thus, rainy pre-harvest weather sets off a chain reaction, in which earlier latent infections are activated by high atmospheric humidity and high soil moisture. High humidity promotes spore production by the fungus from debris and newly-diseased berries, and free water on the berry surfaces promotes germination and new infections by these spores.


Control
Cultural practices such as canopy management and leaf removal are widely practiced and effective components of BBR management programs. Dr. Doug Gubler (University of California, Davis, who first demonstrated the value of leaf removal), emphasizes that this operation should be undertaken during the very early stages of berry development, at fruit set, to help avoid sunburn damage. In encouraging California growers to remove leaves early, Gubler recently noted that some, “have been pulling leaves later and later and are running into sunburn problems, [leading to] secondary fungal infections such as Cladosporium.”

Removal or destruction of vineyard debris, particularly old cluster stems, is theoretically useful and something to keep in mind insofar as it is practical. However, fungicide sprays targeted specifically at BBR may be necessary on susceptible cultivars and/or clones, particularly in a wet year. The questions are, which materials and when?

Not surprisingly, the optimum timing for spray applications is heavily weather-dependent. This probably accounts for the sometimes contradictory opinions and data regarding the relative importance of sprays at the times when they might potentially be applied, such as bloom, pre-bunch close, veraison, and pre-harvest.

A bloom application is designed to prevent the initial establishment of infections through susceptible blossom parts, or infected blossom debris trapped within the cluster (see Photo II), and can be important on susceptible varieties during wet flowering periods, especially if pre-harvest conditions are wet also.

Veraison and pre-harvest sprays are designed to prevent both initial infections through injured berries and, especially, the spread of active infections through the ripening clusters. These are often the most important spray timings in humid climates with regular summer and autumn rainfalls.

Although most effective when applied before an epidemic is in progress, such sprays can still help to slow down the rate of spread if applied relatively soon after the disease first appears and threatens to “take off.”

Unfortunately, most fungicides that control other diseases are relatively ineffective against Botrytis, either providing no significant control or requiring substantially higher rates than required for other diseases. Similarly, most Botrytis-specific fungicides provide no significant control of other disease-causing fungi (including other causes of bunch rot), although there are important exceptions.

Do not forget that all of the Botrytis fungicides are at moderate to high risk of resistance development. This makes it very important to rotate among groups of fungicides used to control this disease, even if you make only occasional applications. Don’t use the same one year after year without substituting something else along the way. Following is a brief review and some new information regarding the fungicides available for Botrytis control.

Before discussing specific materials, let’s define a few of the terms that will be used with respect to fungicial activities:

Protectant — Provides control when applied before the infection process begins.

Post-infection — Provides control when applied for some period of time after the infection process begins, but before symptoms appear. This period of time is often limited to one to several days, but for some materials can be significantly longer with respect to control of latent infections.

Anti-sporulant — Significantly reduces the production of spores from infected tissues, although other symptoms may appear or persist.

VANGARD, SCALA — These represent two different fungicides in the same class of compounds (the anilinopyrimidine or “AP” fungicides), so “rotating” between them will provide no benefit in terms of resistance management.

These compounds appear to have equivalent properties and provide the same general levels of control at labeled rates, although Scala (not registered for application on grapes in California) has been a bit weaker under high disease pressure in a couple of New York trials. In general, these have been the most consistent performers in our field trials over the years, and provide both protective and post-infection activity. (See Table I.)

In recent tests where we’ve inoculated Chardonnay or Pinot Noir clusters with Botrytis spores at bloom, and then sprayed at veraison (before the latent infections became active), both materials eradicated over 90% of latent infections, even though they had been established almost two months earlier.

(NOTE: The clusters in these experiments were sprayed individually by hand, so coverage was absolutely thorough and superior to what would be expected under commercial conditions.) The resistance risk for the AP fungicides is high, so rotation is very important. These products provide no significant control of any disease other than Botrytis.

ELEVATE — The other “work horse” against Botrytis over the past decade. Elevate is not related to any other fungicide on the market, so is an excellent rotational partner. In New York field trials, we’ve seen the same control whether Elevate was rotated with Vangard or either one was used alone. Conventional wisdom has it that this is a protective fungicide, with no post-infection activity.

In contrast, we have several years’ worth of data to indicate that Elevate does have some post-infection activity, and that it can suppress or eradicate latent infections if applied before the infections start to become active. Elevate provides no significant control of any disease other than Botrytis. Resistance risk is moderate.

PRISTINE — The combination product of a strobilurin (pyraclostrobin), plus a compound representing a new chemistry on grapes (boscsalid). Both components are active against Botrytis (boscalid is the more active of the two), which is good from a resistance-management perspective.

Application rate is important. When used at the rate of 8.5 oz to 10.5 oz per acre it is highly effective against powdery mildew and several other diseases, but control of BBR is only fair under anything more than moderate pressure.

There is no mention of Botrytis on the label at this standard use rate, so the company makes no claim concerning activity against this disease. However, Pristine is labeled for “suppression” of Botrytis at a rate of 12.5 oz per acre, and has provided good to very good results in a limited number of tests that we’ve conducted using this rate. In New York, I would consider 12.5 oz per acre to be marginal under very rainy conditions post-veraison, although it will certainly be much better than applying nothing and may be adequate under marginal pressure.

Pristine has a supplemental label for “control” of Botrytis at 18.5 to 23.5 oz per acre. It has provided very good to excellent control at a rate of 19 oz per acre in several trials that we have run under very rainy conditions. It also has a much broader spectrum of activity against other fungi than do the Botrytis-specific materials mentioned above, and it gave very good control of the numerous non-Botrytis rots that developed in our plots during a very wet September in 2006.

The re-entry interval (without protective clothing) is 24 hours at rates up to 12.5 oz per acre, but jumps to five days at the higher rates.

Botrytis-specific materials will provide some control of other rot organisms by reducing the amount of damaged fruit susceptible to secondary invasion, but these are not the only infection courts open to such invaders. Therefore, a broad-spectrum material such as Pristine will provide “additional” control of non-Botrytis rots via direct effects on such organisms, which is why this is such a good material for general bunch rot control.

Pristine provides both protective and limited post-infection activity, although the post-infection activity has not been as great as that for the APs or Elevate in our tests. The resistance risk for both the strobilurin and non-strobilurin component of this product is high, so rotation is important, although the mixture of two unrelated compounds should help reduce the resistance risk.

Note that Flint is also a strobilurin fungicide, so Pristine and Flint should not be rotated with each other unless a non-strobilurin fungicide is also included in the rotation.

FLINT — An excellent powdery mildew fungicide when used at 1.5 to 2.0 oz per acre, it has provided very good to excellent control of Botrytis when applied at the 3-oz per acre rate labeled against this disease. It is primarily a protective fungicide. It does suppress spore production from subsequently-diseased tissues when applied post-infection. Resistance risk is high.

ROVRAL — This was the only true Botrytis fungicide available in the U.S. for nearly 20 years, so rotation just wasn’t an option. As a consequence, resistance appears to be “not uncommon” in many regions where it was used regularly for many years. However, in locations where use has been much more limited, resistance is much less likely, and the material often performs well. Dr. Gubler reports that resistance to this material has not been a problem in California vineyards.

In the absence of resistance, Rovral is an excellent Botrytis fungicide, with both protective and limited post-infection activities. It should be a good rotational partner where resistance is not a major concern, but should be used with caution otherwise. It provides no significant control of any disease other than Botrytis.

“ALTERNATIVE” PRODUCTS — Although others have, we’ve never obtained any control of Botrytis with Stylet Oil or the various potassium bicarbonate products, and I would be very hesitant to rely on them for this purpose under any sort of disease pressure.

Stylet Oil and the bicarbonates both control powdery mildew, which helps to limit Botrytis by eliminating one injury site where this fungus might enter, but direct activity against B. cinerea is questionable. We have obtained significant control with Serenade, but it has never been as effective as the standard products discussed above.