Abstract
Most California vineyards are eventually attacked by one or more grapevine trunk diseases (Esca or Botryosphaeria-, Eutypa- and Phomopsis diebacks). These fungal pathogens cause chronic infections of the wood, which are slow to develop. The symptoms that follow can take years to become obvious. Prevention is an efficient approach, but requires adoption before symptoms appear. To encourage early adoption of preventative practices, economic analyses simulated their benefits in the table grape cultivar Crimson Seedless. Adoption of delayed pruning or pruning-wound protectants was compared in a young, healthy vineyard (years 3 and 5) versus in a mature, diseased vineyard (year 10). A survey of table and raisin grape growers in California’s southern San Joaquin Valley revealed their usage and perceptions of preventative practices. Also, to help growers with mature vineyards, the economic benefits of vine surgery (also referred to as “trunk renewal”) between years 11 and 15 were examined. Our economic simulations showed prevention is cost-effective, if adopted in young vineyards. However, in spite of acknowledging the impact of trunk diseases, only 25 to 30% of growers use preventative practices, and only half of such growers adopt in young vineyards. Further, growers who use prevention and adopt early also perceive preventative practices as more cost-effective. Therefore, an outreach strategy to convince non-adopters must emphasize the long-term economic benefits of early adoption of preventative practices, given the inevitable appearance of symptoms at vineyard maturity. Despite the high one-time cost of vine surgery, our economic analyses suggest its adoption is a cost-effective complement to prevention, and thus, it should be integrated into management recommendations for table grape vineyards at 11 to 15 yrs.
The grapevine trunk-disease complex affects all grape-production systems in California (Gubler et al. 2013). The most common trunk diseases are Esca (also referred to as “Measles”) and the dieback-type trunk diseases: Botryosphaeria, Eutypa, and Phomopsis diebacks. The causal fungi infect the wood, where they reside permanently. The dieback-type diseases kill fruiting positions (spurs), resulting in cumulative yield losses (Munkvold et al. 1994), which make vineyards less profitable over time (Kaplan et al. 2016). Esca has more variable effects, ranging from improper ripening to total vine collapse (apoplexy), which are more challenging to quantify as annual economic losses, but which nonetheless impact fruit and wine quality (Calzarano et al. 2004b, 2009, Lorrain et al. 2012). A common outcome is having to replant the entire vineyard prematurely (Siebert 2001, Sipiora and Cuellar 2014). Given the high cost of vineyard establishment, it is critical to reach the break-even point quickly and then maintain productivity. As such, trunk diseases have a major impact on grape production.
Prevention is an efficient approach because the main infection period is thought to be limited to the dormant season. One preventative practice is to delay pruning from December until February or later, when the period of pruning-wound susceptibility (and thus the risk of infection) is lower in California and parts of Europe (Petzoldt et al. 1981, Larignon and Dubos 2000, Weber et al. 2007, Úrbez-Torres and Gubler 2011). Another approach is to apply a protectant after pruning to minimize infection by fungal spores dispersed with rain or wind. From among the few published field trials, the greatest efficacy against the most pathogens was provided by the protectant thiophanate-methyl (Topsin M WSB; United Phosphorus, Inc.) (Rolshausen et al. 2010), which is one of the few fungicides labeled in California for dormant-season use. Topsin M was more effective against pathogens that cause dieback-type trunk diseases (Botryosphaeria-dieback pathogens Diplodia seriata and Lasiodiplodia theobromae, and the Eutypa-dieback pathogen Eutypa lata), than against Esca pathogens Phaeoacremonium minimum and Phaeomoniella chlamydospora (Rolshausen et al. 2010). Field trials from overseas (e.g., Amponsah et al. 2012, Ayres et al. 2017) often test materials/formulations not available in the United States and/or against different fungal species. Therefore, it is difficult to draw from the published research a preventative practice and precise annual timing to achieve optimal efficacy reliably.
Preventative practices are not always used on a preventative basis (i.e., before symptoms appear). This is in part because in California, infected vines do not begin showing canopy symptoms (e.g., dead spurs, stunted shoots, or symptomatic leaves) until years 8 to 10 (Duthie et al. 1991). In year 10, only 20% of vines are expected to show symptoms, and yield losses are minor (Munkvold et al. 1994). Between years 10 and 15, however, the proportion of symptomatic vines increases to 75% and yield losses become obvious. Indeed, most California winegrape growers we surveyed said they adopt preventative practices when the vineyard is ≥8-yrs-old (Hillis et al. 2017). A survey of pest-control advisers (PCAs) showed they more frequently recommend preventative practices in vineyards with high proportions of symptomatic vines, rather than in vineyards with few to none (Hillis et al. 2016). Economic simulations helped put these survey findings into perspective. By modeling different parameters of disease prevention, vineyard age at the time of adoption was found to be critical (Kaplan et al. 2016). Waiting until year 10 to adopt a preventative practice did not add many more years of profitability unless the practice was optimally effective, something we cannot yet control.
Understanding why growers adopt a particular strategy is essential for effective disease management (Gent et al. 2013) and helps identify important leverage points in the design of outreach programs. In this study, we used an integrated approach to examine adoption of two preventative practices in table grapes: delayed pruning and applications of pruning-wound protectants. First, economic simulations of the table grape cultivar Crimson Seedless were used to evaluate under which scenarios prevention is economically feasible. Next, a survey of California growers was conducted in-person at an outreach workshop in the southern San Joaquin Valley, where 91% of California table grapes are grown (CDFA 2017), to examine grower usage and perceptions of preventative trunk-disease management. The survey questions were designed to determine what proportion of growers use preventative practices, when in the vineyard lifespan they adopt such practices, and whether usage and timing are related to grower perceptions of cost-efficacy. If adoption is related to perceptions of cost-efficacy, then identifying ways to clearly articulate the long-term efficacy of prevention early in a vineyard’s lifespan could increase adoption.
Important differences among grape-production systems could mean not all practices used for winegrapes are feasible for table or raisin grapes. California is the leading producer and sole exporter of table grapes in the United States (CDFA 2017). For example, the preventative practice of double-pruning (the first step of which is mechanical prepruning) is used by some winegrape growers (Hillis et al. 2017), but is suited for cordon-trained cultivars (Weber et al. 2007). Most table and raisin grape cultivars are cane-pruned, which cannot be done with most pruning machines. Also, table and raisin grapes have greater pruning costs than most winegrape vineyards, because pruning is done strictly by hand, and their trellising and training practices tend to support much larger canopies. Such costs may offset economic benefits of preventative practices, making the results of our economic analyses for winegrapes, therefore, inaccurate for other types of grapes. Preventative practices are not 100% effective, a common finding in published field trials, and especially so against some of the Esca pathogens (Eskalen et al. 2007, Rolshausen et al. 2010). Therefore, we must consider a means of treating infected (symptomatic) vines. One such practice is vine surgery, also referred to as “trunk renewal” or “trunk surgery,” which has been evaluated in very few studies, against Eutypa dieback (Sosnowski et al. 2011) and Esca (Calzarano et al. 2004a). This involves cutting off the top of the vine, thereby presumably removing all infected wood, and training a new, healthy shoot from the base of the trunk into a new canopy. Our survey of PCAs suggests there is widespread use of vine surgery in mature vineyards in California, in spite of its high cost (Hillis et al. 2016). As such, in this economic analysis for table grapes, we set out to identify the most cost-effective vineyard age at which to adopt vine surgery, either alone or in combination with preventative practices.
Materials and Methods
Survey design
Disease-management decisions are not always made independently by growers, but are sometimes influenced by vineyard managers, PCAs, consultants, diagnosticians, and extension agents. To gather input for survey content from all individuals relevant to disease-management decision-making, we conducted 22 semi-structured interviews with industry representatives in these roles, through contacts with grower groups like the California Table Grape Commission and University of California Cooperative Extension (UCCE) farm advisors. These interviews confirmed the importance of this broader group of decision-makers, so our survey design addressed the fact that not just growers and vineyard managers, but also PCAs and consultants, might be present during the survey and present different answers.
Compared to our previous studies on winegrapes (Kaplan et al. 2016, Hillis et al. 2017), differences in table grape management, as revealed by the semi-structured interviews conducted before the study, necessitated modifications to our survey and economic analyses. For example, double pruning was not included in this study, even though its efficacy has been evaluated against Eutypa dieback in winegrapes (Weber et al. 2007). The first step of double pruning, the mechanical prepruning, is suited to cordon-trained, spur-pruned vines, but a high proportion of table grapes are cane-pruned and done so by hand. As such, we assumed table grape growers (at least those we interviewed) would be unlikely to consider the preventative practice of double pruning. Also, the greater vigor of table grapes than most winegrapes, in general, produces canopies that are too large for hand application of protectants. Our economic analyses, therefore, included the benefits of spray application of protectants, because most table grape growers we interviewed (and possibly most wine-grape growers, in general) said they were likely to use this application method.
The survey was conducted using the Turning Point electronic audience-response system (Turning Technologies) upon invitation to the San Joaquin Valley Grape Symposium, which was organized by UCCE Specialist Matthew Fidelibus (Kearney Agricultural Center, Parlier, CA) in Easton, CA on 8 Jan 2014. For convenience, respondents are referred to as “growers.” Growers answered each question as it was displayed during a presentation to the audience by author Kendra Baumgartner, using a remote, hand-held device (also referred to as a “clicker”) to key in their answers. Between 30 and 50 growers responded to any given survey question. It took ~30 min to go through the entire survey of 17 questions as a group.
An in-person survey was selected over distribution of an online survey, after our previous experiences with both survey formats. An in-person survey of winegrape growers (Hillis et al. 2017) gave us a greater response rate than an online survey of PCAs (Hillis et al. 2016), allowed a longer, more-detailed survey with more questions, and increased the chance that a given respondent would answer all survey questions. A survey with all questions answered is particularly important to examine the relationship between a respondent’s answers to two different questions and to determine trends in responses across respondents. However, this method reached only a limited number of total growers.
The start of the survey included questions about the growers (job type, years of experience, counties where growers work) and the incidence of trunk diseases in their vineyards. It was necessary to ensure our sample of growers only included those who play a direct role in disease management, determined using the early survey question: Do you make or advise others who make trunk-disease management decisions? Survey responses from individuals who responded “yes” to this question were considered as data for our analyses. The main body of the survey focused on two preventative practices, defined in the survey as follows:
Delayed pruning – Prune late in the dormant season (February or later).
Protectant – After pruning, apply protectant as a spray application (with tractor or ATV-mounted spray nozzles) or by hand (with a paintbrush or sponge to cover pruning wounds).
Growers were asked a series of questions for each practice to gather the following information: frequency of use, vineyard age at which they adopt each practice, and their perceptions of the costs and benefits of each practice. The exact month in which the practices were used was not part of the survey; the exact wording shown above is what the growers saw during the survey.
Statistical analyses
All analyses were conducted with R statistical software (ver. 3.2.1; R Foundation for Statistical Computing), using the base, lme4, rethinking, and ggplot2 packages. A Bayesian modeling framework was used to construct and fit statistical models to our data as described (Hillis et al. 2017). Briefly, outcome variables included the following: frequency of practice usage (scale of 1 to 5 ranging from “Never” to “Always”), timing of practice usage (scale of 1 to 4 ranging from “First three years” to “Year 12 or later”), and cost-efficacy ratings (scale of 1 to 5 ranging from “Very ineffective” to “Very effective”). As these variables are ordinal measures, Bayesian ordered logistic, or proportional odds, regressions were modeled for each outcome (McCullagh 1980). These models estimate parameter values for predictor variables and a vector of cutpoints, which represent the probability that an outcome is in a given category or any of the categories below it (Fullerton 2009). For each analysis, there are at least two candidate models: a null model with no predictors, and a model with two dummy variables for each practice. For the model examining the influence of cost-efficacy perceptions on usage and timing of adoption, cost-efficacy perceptions are included as a predictor, along with the practice dummy variables.
For each predictor variable, the corresponding beta coefficients are reported, as they represent the change in the cumulative log-odds of moving to a higher outcome category as a function of a unit change in the corresponding predictor variable. Also reported are the corresponding 0.95 posterior credibility intervals, which represent the central 95% of the posterior distribution. This approach is an alternative to null hypothesis significance testing, which assesses whether a coefficient is significantly different from zero (or that responses between groups are significantly different). To better understand and visualize the results of model fitting, predictions from each analysis are plotted with predictor variables. The resulting plots represent the probability of the outcome variable being in a particular category as a function of the values of the predictor variables.
Economic analyses
To evaluate the economic benefits of adopting preventative practices, either alone or in combination with vine surgery, we performed a simulated economic experiment that calculated the cumulative discounted net returns (CDNR) over a 25-yr lifespan for a cane-pruned Crimson Seedless vineyard in California’s southern San Joaquin Valley. During our semi-structured interviews for the surveys, which were also done to help parameterize the economic analyses, 25 yrs was what most of the people we spoke with agreed upon as a realistic lifespan for a productive and profitable vineyard, in the absence of trunk diseases. All vineyard management costs on a per acre basis were included. Prices, practice costs, and an age-yield profile were derived from a UCCE costs and returns study on table grapes, and also from discussion with UCCE farm advisors (Table 1). As described previously (Kaplan et al. 2016), annual costs and benefits from production over a 25-yr lifespan are estimated using a published budget for the table grape cultivar Crimson Seedless (Peacock et al. 2007), an updated budget (D. McCorkle, unpublished data, 2015), and historic price data gathered from California grape crush reports published annually by USDA–National Agricultural Statistics Service (NASS). Price and cost values are in 2015 dollars and are discounted using a 3% real discount rate.
Data on yield loss due to trunk diseases was estimated as in our previous study on winegrapes (Kaplan et al. 2016), with a previously described yield-loss function that measured yields over time in California winegrape vineyards with trunk diseases (Munkvold et al. 1994). The disease-incidence function comes from a study that mapped out symptomatic and healthy vines over time in winegrape vineyards with trunk diseases (Duthie et al. 1991). Symptomatic vines were those with canopy symptoms, and thus were assumed to be infected by a trunk pathogen. Healthy vines, however, included uninfected vines and asymptomatic, infected vines. Nonetheless, this model is based on how the proportion of symptomatic vines increases over time and how yields decrease, due to trunk diseases rather than the spread of infection throughout a vineyard.
Briefly, the baseline scenario represents production from a vineyard infected with trunk diseases, and neither preventative practices nor vine surgery are considered. We then simulate scenarios across the two preventative practices, with their different additional costs/acre over and above the cost of standard winter pruning, when adopted at different ages (years 3, 5, and 10), and then used annually. These ages are selected based in part on the pruning practices, yield maturity, and disease cycle, with the beginning of dormant-season pruning in year 3, maturity of yield in year 5, and appearance of symptoms in year 10. Rates of disease-control efficacy (referred to as “efficacy”) are assumed at 25, 50, or 75% of pruning wounds protected, selected from published data on the efficacy of delayed pruning and pruning-wound protectants (Table 2). Next, we simulate vine surgery occurring one time, in years 11, 12, 13, 14, or 15 (based on the assumption that symptoms are discovered in the tenth year), for each preventative-practice scenario. Year 15 was considered to be the end of the optimal age for vine surgery: preliminary analyses showed that adoption in years 16 and up was not cost effective under any scenario. It was assumed that, in March of the year of vine surgery, vines identified as symptomatic the previous growing season are chainsawed off at the base of the trunk, and during the immediately following growing season, one healthy basal shoot is able to be trained into what is assumed to be a new, uninfected vine. In reality, some proportion of symptomatic vines may have infections at their base, and thus the basal shoot grows from infected wood. We have no empirical data to model this; selecting some proportion of vines that are not actually healthy would have been arbitrary. Retrained vines are assumed to have no yield for the first two years (including the year in which vine surgery is performed and the following year), but return mature yields in the third year after surgery, based on discussion with growers. With only two published studies on the efficacy of vine surgery against trunk diseases, Eutypa dieback (Sosnowski et al. 2011) and Esca (Calzarano et al. 2004a), there were no empirical data to estimate efficacy levels. In the simulation, vine surgery was performed on all symptomatic vines in one year, and once they return to mature yields, the cycle of disease development (i.e., the increase in symptomatic vines and the concomitant decrease in yields) starts anew in the vineyard.
Scenarios differ by the cost of pruning-wound protectant application and vine surgery, practice efficacy, and in which years the preventative practices and vine surgery are performed. Delayed pruning is assumed to have no cost in addition to that of standard winter pruning. In other words, labor is assumed to cost the same for pruning, regardless of when during the dormant season pruning is performed. From discussions with growers on possible increased labor costs for pruning crews hired late in the dormant season, we could not get a firm figure. The protectant cost of $64/acre/yr is estimated for a mixture of the fungicides Topsin M and myclobutanil Rally, which also includes the labor cost of spray application with a tractor. Topsin M is the only protectant examined in multiple published studies (Rolshausen et al. 2010, Amponsah et al. 2012), the results of which suggest it is effective against the broadest range of trunk pathogens. Rally is included, not based on its efficacy in studies, but based on discussions with growers, who say they mix it with Topsin M to hopefully achieve a greater level of control against two Esca pathogens, a finding that has not been tested widely or published. Neither fungicide has been tested with spray applications; published studies are for hand applications with a paint brush to individual pruning wounds. With vine surgery, there is a one-time increase in management costs, which we calculate as 400% greater than annual pruning costs from the cost and returns studies, scaled to the percentage of an acre that is removed.
As described previously (Kaplan et al. 2016), the bio-economic model is altered when preventative practices are adopted, and reflects the change in disease incidence (specifically measured as the percentage of symptomatic vines) by reducing the increasing percentage of symptomatic vines and restarting the time step to reflect the new path. For vine surgery, when it is adopted in year τ, the new per acre yield function for the uninfected vines becomes Eq. 1where Removed represents the share of vines removed during vine surgery, and Yield denotes yield for a mature vineyard in year τ. After τ, because we assume vine surgery removes infections from retrained vines (which represent all symptomatic vines in the vineyard in year τ), yield declines as if a new path is restarted, as described above for preventative practices. After two subsequent growing seasons, we assume the retrained vines produce at the mature yield level, As such, after τ + 3, the yield from the retrained and now mature vines starts to decrease as the percentage of symptomatic vines grows over time. We assume that even though vine surgery removes all infected wood in year τ, new pruning wounds made each year are at risk of infection. Note this means, , and
Eq. 2
Results
Economic benefits of preventative practices
Adopting either of the preventative practices alone imposes only a small cost over and above that of standard pruning in December: $0 for delayed pruning and $64/acre/yr for a pruning-wound protectant in the form of a spray application of a mixture of Topsin M and Rally. As such, in our economic simulations, the CDNR are very similar under all scenarios (i.e., under all combinations of years of adoption and disease-control efficacies) of adopting either delayed pruning (Table 3) or pruning-wound protectants (Table 4), especially given our assumption of the same rates of efficacy for both practices. With absolutely no trunk-disease management adopted (neither a preventative practice nor vine surgery), our economic analyses predict a vineyard would be managed at a loss, with a CDNR of -$53,986 after 25 yrs (Tables 3 and 4). This hypothetical diseased vineyard is mainly included for comparative purposes. It seems unlikely a grower would maintain such a vineyard long past year 11, when it is expected to stop generating a profit (Table 5).
The adoption of preventative practices is assumed in our model to lower the rate of cumulative yield losses, which in the vineyard are due to the chronic nature of the numerous infections established through different pruning wounds by different trunk pathogens in different years. Indeed, compared to a negative CDNR under no management practices, we estimate higher CDNR after adoption of preventative practices alone under all scenarios (Tables 3 and 4). Achieving a positive CDNR with adoption of preventative practices alone requires that a preventative practice is adopted in years 3 or 5 (except under 25% efficacy), or in year 10, but only if efficacy is 75% (Tables 3 and 4). These findings of higher CDNR when adopting early suggest that it is best to not wait, especially because efficacy is assumed. Furthermore, CDNR values for adopting in years 3 and 5 are more than twice that of adopting in year 10. Adopting early increases the last profitable year by two (assuming 25% efficacy) to 14 yrs (assuming 75% efficacy). In comparison, adoption in year 10 increases the last profitable year by as few as zero (25% efficacy) and as many as seven (75% efficacy) years (Table 5).
Grower demographics and perceptions of disease severity
Our sample of 54 survey respondents (“growers”) included a high proportion of table and raisin grape growers that were owners and/or PCAs, many of whom farm primarily in Fresno County, which is not surprising given that is where the meeting was held (Table 6). Based on their answers to questions at the start of the survey, survey respondents were relatively balanced in years of farming experience and vineyard acreage farmed (Table 6). Thus, despite a fairly small sample size, we anticipate our findings are roughly representative of the broader industry. The response rate, albeit for these growers who attended the meeting in the first place, was high (45% of people who were handed a clicker).
To understand grower perceptions of the degree of severity of trunk diseases in the vineyards in which they work, we asked growers to give us their best estimate of the percentage of vines infected (i.e., with canopy symptoms) and percentage yield loss due to trunk diseases in their vineyards. In this way, growers acknowledged that trunk diseases are present in their vineyards and impact yields. Nonetheless, they did not report high incidence of either. The median level of infected vines was 5%, with all growers reporting 30% infection or less, except for two reports of very high disease incidence (60 and 100%). The median decline in yield due to trunk diseases was 10%, with all growers reporting 20% or lower, except for three reports of high yield loss (30, 40, and 50%).
Usage and timing of preventative practices
A minority of growers used either delayed pruning or pruning-wound protectants; only 23% of respondents used a preventative practice either often or always (Figure 1A). In contrast, between 50 and 60% of growers either never or rarely used these practices. We also asked growers the typical age of a vineyard when they first started preventing trunk diseases. For both practices, 35 and 48% of growers started to use delayed pruning or protectants, respectively, in vineyards up to 7-yrs-old (Figure 1B), before the symptoms of trunk diseases usually appear (Duthie et al. 1991). In contrast, 65 and 52% of growers start to use delayed pruning or protectants, respectively, in vineyards 13+ yrs-old, which is very late in the disease cycle, when symptomatic vines are widespread and yield losses are apparent (Munkvold et al. 1994). Among growers who timed adoption of either practice before symptoms appear in years 0 to 3, twice as many adopted protectants than delayed pruning (22 vs. 10%, respectively; Figure 1B).
Perceptions of efficacy
More growers rated the practices positively than negatively; when asked about the practice’s ability to maintain yields: 60% of growers rated delayed pruning as somewhat to very effective, versus 20% rating it as somewhat to very ineffective (Figure 2). Protectants were viewed slightly more positively: 68% of growers rated protectants as somewhat to very effective for maintaining yields, vs. 12% rating them as somewhat to very ineffective. When questioned on the cost-efficacy of preventative practices, rather than their ability to maintain yields, fewer growers had positive perceptions: 37 and 50% of growers rated delayed pruning and protectants, respectively, as somewhat to very cost-effective.
Usage and timing as a function of perceptions of cost-efficacy
There was a strong effect of perceived cost-efficacy on the probability of adoption (β = 0.89, CI: 0.47, 1.30): as the cost-efficacy rating improves, the probability of adoption increases. For example, just over 60% of growers who perceived a practice as very ineffective never used it, while practices perceived as very effective were used often or always by >60% of growers (Figure 3A). There was also a substantial effect of cost-efficacy ratings on the timing of adoption (β = -0.40, CI: -0.90, -0.11). For example, 80% of growers who perceived a practice as very ineffective also adopt the practice in mature, diseased vineyards ≥8-yrs-old (Figure 3B). In contrast, 25% of growers who perceived a practice as very effective also adopt the practice in vineyards 0 to 3-yrs-old, thereby using a preventative practice on a preventative basis.
Economic benefits of vine surgery
If we consider adoption of vine surgery alone, our finding of positive CDNR for adoption in any of the five years examined (vineyard ages 11 to 15) suggests that it may outperform preventative practices alone, under some scenarios (Tables 3 and 4). Vine surgery, especially when adopted earlier, may outperform preventative practices when efficacy of the latter is only 25%, based on negative CDNR for preventative practices adopted in years 3, 5, and 10 (Tables 3 and 4). If vine surgery alone is used to manage trunk diseases, our findings of a $22,728 difference in CDNR between adopting in year 11 versus year 15 suggest there is a substantial cost of waiting until year 15. Yield losses typically build exponentially during these five years (Munkvold et al. 1994), as do the proportion of symptomatic vines (Duthie et al. 1991), so each year, the one-time cost of performing surgery increases with the increasing number of symptomatic vines, and there are more lost revenues from their declining yields.
Lastly, if we consider combining a preventative practice (either delayed pruning or applications of pruning-wound protectants) with vine surgery, our economic analyses reveal higher CDNR (especially under low efficacy) than when either a preventative practice or vine surgery is used alone. CDNR varies negligibly regardless of when vine surgery is adopted between years 11 and 15, especially with a preventative practice adopted early in years 3 or 5, and with 75% efficacy (Tables 3 and 4). However, a common situation in table grape vineyards, based on our survey results, is that no preventative practices are adopted until year 8 and up (Figure 1B). Indeed, when adopting a preventative practice in year 10, addition of vine surgery (even with only 25% assumed for efficacy of the preventative practice) is associated with CDNR values that are more than twice that of the same preventative practice alone (Tables 3 and 4). However, when following adoption of a preventative practice in year 10, the year in which vine surgery is adopted is associated with much greater variation in CDNR, compared to adoption in years 3 or 5. For example, if a preventative practice is adopted in year 10, regardless of its efficacy, addition of vine surgery in year 11 versus year 15 is associated with a decrease in CDNR by between 13 and 24%. In comparison, if a preventative practice is adopted in years 3 or 5, the addition of vine surgery among the years examined is associated with relatively negligible differences in CDNR.
Discussion
Our findings of low adoption rates, assuming the table and raisin grape growers who attended our meeting in the southern San Joaquin Valley of California can represent the population of all of such growers, suggest the majority of growers do not routinely prevent trunk diseases. A minority of growers said they delay pruning or apply protectants, but only after the vineyard is mature (≥8-yrs-old). Based on what is known about the disease cycle, symptomatic vines are present in the vineyard at age eight and older (Munkvold et al. 1994). A preventative practice adopted late in the disease cycle has no effect on infections initiated in previous years. Because we sampled all growers who attended our meeting, our sample is not random with respect to the overall population of table and raisin grape growers in the southern San Joaquin Valley. We believe our findings are still relevant, however, in part because survey respondents represented comparable proportions of growers farming small, medium, and large vineyards. Also, that these growers attended the meeting in the first place suggests that they stay informed and view participation in extension activities as important. Therefore, we might expect them to be more likely than average to be receptive to the idea of prevention. Thus, overall rates of adoption in the general population are plausibly lower than we report here.
Based on our finding that ~25% of survey respondents adopt either delayed pruning or protectants in vineyards ≤3-yrs-old, it seems there are at least a few growers who use preventative practices in a truly preventative manner. Among these growers, delayed pruning was used less often than protectants. This finding is in contrast to our winegrape study, in which delayed pruning was a more common preventative practice than protectants or double pruning (Hillis et al. 2017). The risk of delaying budbreak and, subsequently, harvest (which could shift the timing of getting fruit to market), may prevent table grape growers from adopting this preventative practice widely. Alternatively, the labor-intensive nature of pruning table and raisin grapes, which is done exclusively by hand, makes securing a pruning crew a top priority. Delaying pruning for trunk-disease management (or for any other reason) may be of less importance when scheduling pruning. In contrast, applying a pruning-wound protectant requires just a tractor driver, making this practice easier to schedule than delayed pruning.
Our finding that a high proportion of surveyed growers adopted preventative practices in mature, infected vineyards may be due in part to the long-term nature of the disease cycle, which makes it difficult for both growers and researchers to see the results of adopting a preventative practice. Interestingly, the same growers who said they do not adopt a preventative practice until the vineyard is ≥8-yrs-old also had negative perceptions of practice efficacy and, not surprisingly, do not think these practices are cost-effective. We assume that the multi-year delay between infection and symptom development confounds the process of learning about preventative practices on a trial-and-error basis (i.e., “experiential learning”). Growers naturally try out new practices when they are farming. In deciding whether a practice is worth repeating, they reflect on their past experiences, recall whether the practice was effective, and adjust accordingly. However, because of the lag between infection and symptom expression, experiential learning can lead them astray. They may believe they are using a practice in a preventative manner, but they are actually using it after infection has occurred and before symptoms are widely expressed throughout the vineyard. After using the practice, they still see symptomatic vines (possibly more than before using the practice). The net result of this process of experiential learning is a negative perception that trunk disease preventative practices are not really worth it. This scenario is of course speculative, without knowing exactly why survey respondents chose to adopt a preventative practice in a mature vineyard, where we assume symptomatic vines were already present. Nonetheless, we have consistent findings from our winegrape study: among the few growers who use prevention, most adopt practices too late in the disease cycle and a large proportion of those who do so perceive preventative practices as ineffective (Hillis et al. 2017).
Our economic simulation model suggests that, under market conditions like those found in the southern San Joaquin Valley, when table grape growers adopt a preventative practice and use it annually to treat trunk diseases, they should expect the treatment to generate positive net benefits, unless adopted after symptoms appear (as is typical in a mature vineyard) or if its efficacy is low (something we cannot control). Growers should also consider vine surgery after symptoms appear in the vineyard, but when the vineyard is close to 11-yrs-old. The earlier a preventative practice is adopted, the later vine surgery is necessary, as a low proportion of symptomatic vines means it is more cost-effective to delay vine surgery. Early adoption of preventative practices can more than double the profitable lifespan of a vineyard, especially when the treatment is highly effective.
Vine surgery, defined in our economic analyses as retraining a new vine from the base of the trunk (rather than just retraining a new cordon), is more costly than preventative practices. Estimates provided by PCAs (Hillis et al. 2016) and based on our own calculations put the one-time cost of vine surgery as at least $1340/acre, or four times the annual cost to prune the vineyard ($335/acre), and 21 times the annual cost of a spray application of a pruning-wound protectant, such as a mixture of Topsin M and Rally ($64/acre). Nonetheless, our analyses suggest that adding vine surgery to early adoption of a preventative practice generates even greater net returns for a table grape vineyard than adopting either approach alone. Vine surgery is also expected to extend the profitable lifespan of the vineyard to 25 yrs (the maximum considered in our analyses), when preventative practices turn out to have low efficacy, and thus fall short of the 25-yr mark. In the long run, the combination of early adoption of preventative practices and vine surgery provides for a more profitable and sustainable vineyard.
Vine surgery has undergone far less testing than preventative practices: only two studies evaluated the efficacy of vine surgery against Eutypa dieback (Sosnowski et al. 2011) and Esca (Calzarano et al. 2004a). Furthermore, we cannot assume similar efficacy against all trunk diseases, especially Esca, the wood symptoms of which can be found below the graft union and extending into the rootstock (Gramaje et al. 2010). This makes it difficult during vine surgery to cut out what is thought to be infected wood, because Esca pathogens are less likely than dieback-type pathogens (Eutypa lata, Diaporthe spp., or Botryosphaeriaceae) to be limited to within the margins of the visible wood symptoms. Esca is different from dieback-type trunk diseases in that some of the Esca pathogens are culturable far beyond the characteristic black streaking in the wood (Pouzoulet et al. 2013), and also have been isolated from vines with no wood symptoms (Travadon et al. 2016). If wood infected by Esca or other trunk pathogens is present below the graft union, then the new vine that is to be trained from just above the graft union will not be healthy, and the costs of vine surgery may not be offset by a return to productive crop yields.
Conclusions
Our economic simulations assume efficacy rates of 25, 50, or 75%, but this is obviously not a parameter we can control in the field. Very little is known about the climate conditions that are optimal for sporulation and germination, beyond the fact that rain is important for spore production and, for some trunk pathogens, also for spore dispersal. Nonetheless, we made “blanket” assumptions about practice efficacy across trunk diseases. Theories on pruning-wound susceptibility come from a handful of studies on Eutypa dieback (Petzoldt et al. 1981, Weber et al. 2007) and Botryosphaeria dieback (Úrbez-Torres and Gubler 2011). However, the period of pruning-wound susceptibility may vary in different climates and in different years, further influencing efficacy. For example, if rainfall is atypically low between December and February, disease risk may be low, and applying pruning-wound protectants during this time (or delaying pruning until after this time) may not be necessary. Spore-trapping studies show very few spores in dry winters, even after it does rain (Úrbez-Torres et al. 2010). Because little is known about pruning-wound susceptibility, our best recommendation is to choose a preventative practice and use it routinely each year, starting when the vineyard is young. Follow with vine surgery when the proportion of symptomatic vines is ~20% or lower, to justify the large one-time cost.
Acknowledgments
The authors thank Dr. Matthew W. Fidelibus (Department of Viticulture and Enology, University of California, Davis) for organizing the workshop at which the survey was conducted and Dr. Dean McCorkle (Texas A&M Agrilife Extension) for providing an updated table grape cost and return study. This research was funded by grant 2012-51181-19954 to K. Baumgartner, J. Kaplan, and M. Lubell from the USDA, National Institute of Food and Agriculture’s Specialty Crops Research Initiative. Mention of trade names or commercial products is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA. USDA is an equal opportunity provider and employer.
Footnotes
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. The full statement of the Disclaimers is available at http://www.ajevonline.org/content/proprietary-rights-notice-ajev-online. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received September 2018.
- Revision received December 2018.
- Accepted January 2019.
- Published online July 2019
- ©2019 by the American Society for Enology and Viticulture