Using ontogenetic resistance of grapevine for fungicide reduction strategies

Resistance of grapevine against powdery mildew (Erysiphe necator) and downy mildew (Plasmopara viticola) increases during ontogenetic development of leaves and berries. The aim of this study was to determine whether the use of fungicides against both pathogens could be reduced during the last treatment before harvest (LTBH) by treating only the susceptible leaves of the vines in the upper half of the canopy. For this purpose, field trials were carried out under experimental or practical conditions in vineyards in southern Germany in eight different years. All plants were sprayed after a standard regimen until LTBH, performed at BBCH 77–79. LTBH was either performed on the whole canopy or limited to the upper half of the canopy excluding the fruit zone. For both pathogens, no differences were observed between the two treatments with regard to the disease severity on the grape clusters. Only in one out of eight years, an increased fruit disease incidence could be observed. As far as the disease severity of leaves is concerned, no differences were observed for downy mildew, and in the case of powdery mildew, significantly higher infestation was observed only in two out of eight years. Furthermore, the efficacy of different fungicides for LTBH was investigated in a 3-year experiment. Sole use of a copper fungicide for LTBH had no disadvantages against E. necator, compared to a treatment with synthetic fungicides. Taken together, LTBH at BBCH 77–79 can be limited to the upper canopy without risk to the health of the fruit.


Introduction
Downy mildew (Plasmopara viticola) and powdery mildew (Erysiphe necator) of grapevine regularly cause great damage in vineyards around the world. Usually, both pathogens are controlled by regular use of fungicides, with particularly high number of applications in humid environments (Pertot et al., 2016). The extensive use of pesticides and industrialized agriculture in general are increasingly associated with negative environmental changes such as species extinction (Raven & Wagner, 2021). As a result, plant protection has become a topic of political and social discourse, and Europewide regulations such as the National Action Plans for the Sustainable Use of Plant Protection Products as well as country-specific legislative amendments increasingly demand the implementation of the principles of integrated pest management and a consistent reduction of the plant protection products used (BMEL, 2013). In recent decades, a reduction of the total amounts of applied fungicides has been achieved due to breeding of resistant cultivars (Pedneault & Provost, 2016), optimized application techniques (Gil et al., 2014) or the availability of decision support systems .
While classic grapevine cultivars are highly susceptible during most of the season, their resistance increases in maturing grapevine organs. Ontogenetic or age-related resistance has been described for the most important fungal diseases of grapevine (Deytieux-Belleau et al., 2009; (Ficke et al., 2002). Disease severity already decreases rapidly on clusters inoculated twelve days after bloom (Kast & Stark-Urnau, 2000). With berries beginning to touch (BBCH 77), three to four weeks after bloom, berries are almost completely resistant (Gadoury et al., 2003;Kast & Stark-Urnau, 2000). In the case of downy mildew, grape clusters are susceptible as soon as inflorescences become visible (BBCH 53) and their susceptibility decreases few weeks after bloom (Kennelly et al., 2005). Increasing resistance was also observed in maturing leaves. While in greenhouse experiments the first fully developed leaf beneath the apex of the plant showed a disease severity of 90% after inoculation with P. viticola, the next two leaves showed only about 80% and 50%, respectively (Steinmetz et al., 2012).
Regardless of the enhanced resistance of older leaves and clusters at the end of the season, plant protection treatments against powdery mildew and downy mildew are usually carried out until just before the onset of ripening. In the case of powdery mildew, modern spray programs usually take into account ontogenetic resistance (Kast & Bleyer, 2011;Kast & Schiefer, 2003). Thereby the most potent agents can be used during the critical phase, facilitating an effective resistance management through frequent changes of active ingredients (Gubler et al., 1999;Moyer et al., 2018). By using index values which calculate the general risk of powdery mildew, the duration of action of the agents can be estimated and the spraying intervals can be planned (Dubuis et al., 2019). In the case of downy mildew, plant protection treatments are mainly timed according to new grown leaf area and the amount of rain since the last treatment (Bleyer et al., 2020).
The aim of this study was therefore to evaluate whether the enhanced resistance of maturing grape organs could be utilized to reduce fungicide use during the final treatments of the season. Fungicide applications were therefore restricted to the upper half of the canopy and evaluated for their sufficiency to harvest healthy fruits. Since treatments against powdery mildew and downy mildew in practice are usually carried out simultaneously with the same tank mixture, the feasibility of the presented approach for both pathogens was investigated. The results of this work prove that the importance of the last treatment at approximately BBCH 77-79 (berry touch) is often overestimated in practice. However, since the last treatment before harvest (LTBH) is usually carried out at least six weeks before harvest or earlier, it may be important to maintain the leaves of the upper canopy as an additional assimilation area to ensure adequate fruit and wood maturity. By implementing the recommendations and procedures described in this publication, winegrowers can benefit from some ecological and economic advantages e.g., reduced amounts of plant protection products with consequent lower residues on fruits and lower costs for fungicides.

Material and methods
Experimental conditions for precise evaluation of efficacy against Erysiphe necator The efficacy of a last treatment before harvest (LTBH) against powdery mildew restricted to the upper half of the canopy excluding the fruit zone was evaluated in six years (2009, 2010, 2012, 2013, 2014 and 2016) at the State Education and Research Institute for Viticulture and Pomology (LVWO; Weinsberg, Germany). The experimental vineyard (49°09′25.5"N 9°18′29.8″E) was planted with 4,150 highly susceptible Vitis vinifera cv. Cabernet Dorsa plants / ha (row spacing 2 m, plant spacing 1.2 m) in Espalier-type system on loamy Keuper soil. Untreated separation rows located between the rows of the different variants were artificially infected to ensure a high and uniform infestation with powdery mildew. For this purpose, potted greenhouse plants infected with E. necator were placed in the vineyard next to every seventh vine in the separation row. Plant protection applications were carried out in every single row, except of the separation rows, with a tunnel-sprayer (Schachtner Geraetetechnik, Ludwigsburg, Germany) with TeeJet 6501 (≤ 600 L/ha) and 65015 (800 L/ha) nozzles (TeeJet Technologies, Schorndorf, Germany), respectively. Until LTBH at approximately BBCH 77-79 (berry touch), all plants were sprayed after a standard schedule against powdery mildew with adjusted application rates according to the phenological state from the basic application rate at BBCH 16 (1x concentrated in 400 L/ha) to 3.5x basic application rate at BBCH 73 (1.75x concentrated in 800 L/ha) (Supplemental data 1). Treatments were timed accordingly to the decision support system VitiMeteo-OiDiag (https://www.vitimeteo. de; Kast & Bleyer, 2014). LTBH was either applied to the whole canopy or limited to the upper half of the canopy, excluding the fruit zone, by closing the lower three of the five nozzles of the sprayer. LTBH was performed with a fungicide of the demethylation inhibitors group (DMIs) as recommended by the official German viticulture advisory services. Selected were therefore the commercially available triazole fungicides Systhane™ 20 EW (200 g/L myclobutanil; Corteva Agriscience, Germany) in 2009, 2010, 2012, 2014 and 2016 and Topas® (100 g/L penconazole; Syngenta GmbH, Germany) in 2013, respectively. To get an impression of the severity of the epidemic in the individual trial years, a control that was not treated against powdery mildew (untreated control) was also integrated. Valuation was carried out after the European and Mediterranean Plant Protection Organization (EPPO) standards in four repetitions per treatment in the randomized vineyards. The severity of the disease was rated for both experimental conditions (LTBH on whole canopy and LTBH on upper half of the canopy) and the untreated control by visually determining the percentage of symptomatic leaf/cluster surface area on 4 × 100 leaves or clusters between BBCH 83 (berries developing color) and BBCH 85 (softening of berries). The disease incidence was calculated by dividing symptomatic leaves or clusters by the total number of leaves or clusters examined. Statistical analysis for significance was performed by one-way analysis of variance (ANOVA; p = 0.05) with Student-Newman-Keuls (SNK) method as a post-hoc test.
Experimental conditions for precise evaluation of efficacy against Plasmopara viticola The efficacy of LTBH against downy mildew restricted to the upper half of the canopy excluding the fruit zone was evaluated in six years (2009, 2010, 2012, 2013, 2014 and 2016) at LVWO (Weinsberg, Germany). The vineyard (49°09′18.0"N 9°16′37.7″E) was planted with 4,150 V. vinifera cv. Riesling plants / ha (row spacing 2 m, plant spacing 1.2 m) in Espalier-type system on Keuper soil. Untreated separation rows located between the rows of the different variants were artificially infected to ensure a high and uniform infestation with downy mildew. For this purpose, two leaves from a single shoot of every eighth vine were inoculated by spray infection with a sporangia solution of a P. viticola isolate from LVWO. To ensure the viability of the isolate, it was refreshed with collected sporangia every year. The sporangia solution was produced by rinsing of infected leaves with desalted water and was adjusted to a concentration of 25,000 sporangia/mL. Infection was achieved with a commercially available pump-sprayer on a spot that was 3 cm in diameter on the lower side of the leaf. Plant protection treatments were timed accordingly to the decision support system VitiMeteo-Plasmopara (https://www.vitimeteo.de/; Bleyer et al., 2008) and performed after a standard regimen against grapevine downy mildew in the manner mentioned above (Supplemental data 1). LTBH at approximately BBCH 77-79 (berry touch) was performed either on the whole canopy or only on the upper canopy as for powdery mildew but with the commercially available synthetic fungicide Folpan® 80 WDG (800 g/kg folpet; Adama GmbH, Germany). To get an impression of the severity of the epidemic in the individual trial years, a control that was not treated against downy mildew (untreated control) was also integrated. Evaluation was carried out between BBCH 81 (beginning of ripening) and BBCH 85 (softening of berries) after EPPO standards as mentioned above.

Experimental setup under standard practical conditions
Trials under practical conditions were performed on the cultivar V. vinifera cv. Pinot gris during the years 2014, 2015, 2016 and 2017 in a vineyard (48°02′42.0"N 7°37′ 17.3″E) of a winery in Ihringen (Germany). Plant protection treatments were performed in every second row with a trailed Myers-sprayer (Hans Wanner GmbH, Wangen im Allgaeu, Germany) with axial-blower and yellow Albuz® ATR80 nozzles by the personnel of the winery. Until LTBH at approximately BBCH 77-79 (berry touch), all vineyards were sprayed after a standard schedule against powdery and downy mildew as mentioned above. LTBH was performed with the commercially available products Folpan® 80 WDG (800 g/kg folpet; Adama GmbH, Germany) and Systhane™ 20 EW (200 g/L myclobutanil; Corteva Agriscience, Germany) in two variants either on the whole canopy (five of five nozzles open) or only on the upper canopy by closing of the lower three nozzles. Valuation of the plots of the two experimental conditions (LTBH on whole canopy and LTBH on upper half of the canopy) was carried out by officially trained personnel from State Institute of Viticulture and Enology (WBI; Freiburg, Germany) using EPPO standards as mentioned above.
Comparison of synthetic fungicides vs. copper during the final treatment Precise evaluation of the efficacy of a sole copper application during LTBH was performed in a vineyard of the State Institute of Viticulture and Enology (WBI; Freiburg, Germany). The vineyard (47°58′38.6"N 7°50′12.4″E) was planted with 5,000 V. vinifera cv. Chasselas plants / ha (row spacing 2 m, plant spacing 1 m) in Espalier-type system on loam soil. Plant protection treatments were performed in every single row with a tunnel-sprayer (Schachtner, Ludwigsburg, Germany) with TeeJet XR80015VS nozzles (TeeJet Technologies, Schorndorf, Germany). Until LTBH at approximately BBCH 77-79 (berry touch), all plants were sprayed after a standard schedule against powdery and downy mildew (Supplemental data 1). LTBH was performed either on the whole canopy (five of five nozzles open), only on the upper canopy by closing of the lower three nozzles or omitted completely. LTBH on the whole canopy was performed with the combination Folpan® 80 WDG (800 g/kg folpet; Adama GmbH, Germany) and Topas® (100 g/L penconazole; Syngenta GmbH, Germany). LTBH on the upper canopy was performed in three different variants, either with the combination Folpan® 80 WDG and Topas®, the combination Cuprozin® progress (383.8 g/ L copper hydroxide; Spiess-Urania Chemicals GmbH, Germany) and Topas® or with Cuprozin® progress only. Evaluation of the five variants was carried out after EPPO standards as mentioned above.

Results
Between 2009 and 2017 a total of 19 field trials were carried out under several experimental conditions to analyze the potential for fungicide reduction during LTBH of grapevine powdery and downy mildew control. All plant protection applications were carried out according to a standard regime and varied only in the last treatment.
LTBH restricted to upper canopy is sufficient for harvest of healthy fruits Efficacy evaluation of a reduced LTBH was performed individually for downy mildew and powdery mildew by precise trials in experimental vineyards of LVWO. LTBH was applied either on the whole canopy as usual or only on the upper half of the canopy. An untreated control was included to monitor the severity of the epidemic. Differences between both treatments were only detected in years with particularly high infection pressure.
With exception of 2009, when infection pressure was so low that only minor differences between untreated control and treated variants were observed, all years showed sufficient powdery mildew infections to estimate the efficacy of the treatments. Leaf disease incidence showed no significant differences regardless of LTBH was applied to the whole canopy or to the upper half of the canopy (Table 1). Leaf disease severity, however, was significantly increased from 4.6% in the whole canopy treated variant to 19.9% in the upper half treated variant in 2016. Differences regarding fruit disease incidence between the two applied variants were only detected in 2012 (whole canopy: 50%, upper half of the canopy: 95.8%). However, by the end of the season all years had produced healthy fruits showing no differences in fruit disease severity, regardless of the characteristic of LTBH.
The results for the precise experiments performed with the pathogen P. viticola showed a similar result (Table 2). For the protection of leaves both treatments were equally sufficient. Differences regarding fruit health were only rarely detected. While in 2013 fruit disease incidence was significantly higher in the upper half variant (14.3%) compared to the whole canopy variant (4.3%), fruit disease severity was not affected. In summary, healthy fruits were harvested in all trial years except 2016. Even though P. viticola was not sufficiently controlled in 2016 due to the unusually heavy epidemic (Bleyer et al., 2020), significant differences on plant health between LTBH on whole canopy (fruit    Exact trials with Plasmopara viticola were performed in a vineyard of the State Education and Research Institute for Viticulture and Pomology (LVWO) and rated at BBCH 81-85. Different letters indicate significant differences (p ≤ 0.05); trt. = treatment; utc = untreated control; wc = last treatment before harvest (LTBH) on whole canopy; uc = LTBH only on upper canopy; inc = disease incidence; sev = disease severity; values after ± show standard deviation disease severity: 13.3%) or upper half of the canopy (fruit disease severity: 16.2%) were not detected. To verify the results obtained from the precise experiments in experimental vineyards of LVWO, trials were also performed with a standard sprayer under practical conditions in a winery. Powdery mildew was so efficiently controlled over the four years of the trial that no infestation of fruits could be assessed in any of the years. Only in 2015 a weak infection on leaves was assessed, but significant differences between the LTBH variants (whole canopy or upper half of the canopy) were not detected. In the case of downy mildew, differences between LTBH on whole canopy or upper half of the canopy were only detected in 2017. Whole canopy treatment performed significantly worse than upper half treatment, but the differences occurred only in fruit disease severity and were only 1.8% and 1.0%, respectively. In 2014 and 2015, the different treatments showed no significant differences (Supplemental data 2).

Copper is sufficient for the control of the two pathogens in LTBH
Since the fruit zone had a relatively high level of resistance to both pathogens at the time of LTBH, the possibility of excluding synthetic fungicides during this step was investigated. Although no positive effect of LTBH on fruits was demonstrated in any of the three years observed (Table 3), LTBH can significantly improve leaf health. Leaf disease incidence and severity of downy mildew was significantly reduced in 2017, however, differences between the fungicide Folpan® 80 WDG on whole canopy or upper half of the canopy or the Cuprozin® progress treatment on upper half of the canopy were not detected.
Against E. necator, the copper fungicide showed a significantly better efficacy than the fungicide Topas® in two out of three years. In 2015 no differences between whole canopy and upper half treatment were detected for the synthetic fungicides. However, treating upper half of the canopy with Cuprozin® progress reduced disease incidence on leaves significantly. A combination of copper with penconazole had no further beneficial effect. 2017 showed a similar result and proved that a sole copper spray on the upper half of the canopy performs at least equally then a treatment with the fungicide Topas® on whole canopy under the given conditions. If both fungicides were applied only to the upper half of the canopy, copper performed even better against powdery mildew then the substance penconazole.

Discussion
Grapevine accounts for the largest proportion of fungicides used in the European Union and the main causes of large-scale fungicide use in viticulture are powdery mildew and downy mildew (Eurostat, 2007). While older leaves and berries of grapevine become more and more resistant to these two diseases during ontogenetic development, younger leaves, mainly present in laterals in the upper canopy, are highly susceptible until harvest (Gadoury et al., 2001;Kennelly et al., 2005). During the six or more weeks between the final plant protection application of the season and harvest, these leaves contribute significantly to fruit ripening and reserve starch storage in the wood, especially in vine growing regions with warm-summer humid continental / maritime climate (Zufferey et al., 2012). Considering these aspects, this work evaluated the possibility of fungicide reduction through targeted application to the upper half of the canopy by closing the lower nozzles of the sprayer during the last treatment before harvest.
Trials in years with high infection pressure at LVWO, such as 2012 for E. necator or 2013 for P. viticola, show that although fruit disease incidence is increased, fruit disease severity is not significantly affected if the fruit zone is excluded during LTBH. Trials performed at WBI between 2015 and 2017 furthermore show that LTBH against powdery and downy mildew at BBCH 77-79 is only necessary in particularly difficult years. Heavy infection pressure like in the year 2017 showed a beneficial effect of LTBH on leaves. While P. viticola showed no differences between whole canopy and upper canopy treatment, leaf disease severity of E. necator was significantly higher in the upper canopy than in the whole canopy variant, albeit on a very low level. Nevertheless, when the fruits were kept healthy until berries touch (BBCH 77-79), a beneficial effect of LTBH on fruit quality was not evident. However, when the last two applications of the season were restricted to the upper canopy, fruit disease severity of P. viticola significantly increased in three out of nine years (data not shown).
Since ontogenetic resistance is mainly explained by the altered morphology of maturing berries and rachis this was unexpected. While P. viticola inevitably requires functional stomata to enter the plant, E. necator penetrates the host cell directly through the epidermis. Microscopic analysis of several cultivars at BBCH 69 (all caps fallen) and BBCH 75 (pea sized berries) showed that at both stages stomata on receptacles, pedicels of berries or rachis are closed or collapsed in increasing numbers compared to BBCH 53 (inflorescences clearly visible, blossom buds not separated) . Also the layer of cuticular waxes increases as the berries develop, potentially providing direct protection of the underlying epidermis as well as indirect protection through poorer spore attachment (Arand et al., 2021;Gee et al., 2008). Nevertheless, due to few remaining functional stomata still present at BBCH 77 (berries beginning to touch), particularly favorable conditions for P. viticola may lead to infection this late in the season. As treatments against both pathogens are carried out simultaneously in practice, two applications excluding the fruit zone at the end of the season are therefore not recommended as a general rule.
Another important point when planning LTBH is the choice of fungicides. While precise evaluation of LTBH at the State Education and Research Institute for Viticulture and Pomology (LVWO) against P. viticola was performed with the phthalimide folpet, experiments with E. necator were carried out with the DMI-fungicides myclobutanil and penconazole, respectively. Folpet was chosen due to its non-systemic characteristics and the low 4.0 ± 2.6 b 0.3 ± 0.5 a 0.0 ± 0.0 a 90.5 ± 6.2 ab 4.1 ± 0.6 c leaves uc (F + P) 53.0 ± 22.7 b 3.2 ± 2.1 b 0.5 ± 1.0 a 0.1 ± 0.2 a 93.5 ± 3.8 a 5.9 ± 0.6 b uc (C + P) 17.8 ± 11.1 c 0.6 ± 0.5 b 12.3 ± 20.5 a 1.4 ± 2.6 a 80.0 ± 9.9 b 3.5 ± 0.4 c uc (C) 21.5 ± 2.6 c 0.9 ± 0.1 b 2.5 ± 5.0 a 0. Exact trials with Erysiphe necator and Plasmopara viticola were performed in a vineyard of the State Institute of Viticulture and Enology (WBI) and rated at BBCH 81-85 to evaluate the effect of sole copper spray as a last treatment before harvest. Different letters indicate significant differences (p ≤ 0.05); trt. = treatment; − = no treatment at timepoint of LTBH; wc = LTBH on whole canopy; uc = LTBH only on upper canopy; F = folpet; P = penconazole; C = copper; inc = disease incidence; sev = disease severity; values after ± show standard deviation risk of resistance development. As already mentioned above, the DMI-fungicides against powdery mildew were chosen due to the current recommendations of the official German viticulture advisory services. LTBH against E. necator with a contact fungicide was unfortunately not possible as sulfur is not permitted that late in the season due to its preharvest interval of 56 days in Germany. However, as myclobutanil as well as penconazol are translocated through the xylem an effect on clusters after LTBH on the upper half of the canopy should not be present (Baibakova et al., 2019). Particularly noticeable from this study was the good effect of a copper fungicide against powdery mildew. The variant treated with the copper fungicide performed better than the variant sprayed with the synthetic powdery mildew fungicide Topas®. A combination of the copper fungicide and penconazole showed no differences to the sole copper treatment. Although the recommendations of the resistance management valid in the corresponding year were followed and the fungicides of the FRAC 3 group were not applied more than four times in the season, it cannot be completely excluded that the poor efficacy of the azoles was due to fungicide resistance in E. necator. The effect of copper against powdery mildews is known since decades (RMoyle & Griffin, 1973). Nevertheless, the use of copper in practice against this disease is often underestimated and synthetic fungicides are recommended in practice. Powdery mildew control in organic viticulture mainly relies on sulphur, which must not be applied at the end of the season to its preharvest interval in Germany of 56 days in fruits used for wine production. New products approved for the control of E. necator in organic viticulture based on Bacillus amyloliquefaciens or chitosan formulations became recently available, however these products have not been tested extensively. Based on the results presented here, copper is a low cost solution during LTBH providing excellent control of both pathogens with a preharvest interval of only 21 days. New formulations of copper products with slow and continuous release of the active agent, providing a better performance on grape clusters than the currently available products, may furthermore give the possibility to limit more applications to the upper canopy (Weitbrecht et al., 2021).
Another benefit of copper may come from the inhibition of E. necator chasmothecia formation and the reduction of inoculum in the following season (Redl et al., 2021). This is particularly crucial since chasmothecia are, next to flag shoots developed from dormant infected buds, the most important source of infection in the following season (Ruegner et al., 2002). Nevertheless, E. necator in particular can spread undetected in the vineyard for a long time, which is why the detection of the disease outbreak and the right timing of the first treatments during the season are much more important than the last treatments before harvest.
It is also important to emphasize that plant protection treatments must always be carried out after individual consideration and assessment of the existing conditions. While Gadoury et al. (2003) found no cultivar dependent influence, the onset of ontogenetic resistance is significantly depending on the location of the vineyard, climate and consequently plant development. Kennelly et al. (2005) showed for example, that berry susceptibility is significantly prolonged at a site with mild winters resulting in a longer flowering period compared to a site with cold winters and a shorter flowering period. Especially in times of rapidly changing climatic conditions, one must always keep in mind the possible effects of changing temperatures on the morphological development of the vine organs and consequently the impact on ontogenetic resistance. Winemakers should therefore always consider the current development of the epidemic before making an individual decision on the necessity of a final treatment.
Besides the positive effects on the environment, another possible advantage of LTBH only on the upper half of the canopy may be lower residues of plant protection products on fruits. Analysis of grape must and wine for residues of the pesticides used showed a reduction in the grape must, but the levels of all variants, regardless of spray treatment, were far below the legal limits or near the detection limit in the wine. However, these analysis was only performed in two years and two technical replicates per variant (data not shown).
Taken together, the present study shows a practicable way to significantly reduce the amount of fungicide used. Even if the savings of half a treatment at the end of the season seem small at first glance, the potential savings on a larger scale are quite considerable. Halving of the fungicide amount during LTBH with penconazole and folpet would result in a saving of 24,000 kg per year for the~25,000 ha of vines in south-west Germany. At the current average price, a winegrower could save 14.60€ per hectare with the folpet / penconazole combination, which would correspond to 365,000€ per year in south-west Germany. For the significantly more expensive copper fungicide, which is used in organic viticulture, the absolute quantities and consequently the monetary advantage would be even higher. By implementing the strategy presented here, winegrowers benefit therefor through saving money, less impact on the environment, and consequently a better reputation in the public mind.

Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s10658-022-02592-w. Heger (Ihringen, Germany) for the opportunity to validate our results on their estate and Jan-Dominik Schneider and Thomas Amann for the excellent execution of the spraying work at the winery. We are also especially grateful to René Fuchs for critical reading of the manuscript. Our very special thanks go to Walter Kast for the original ideas leading to the experiments regarding the last treatment before harvest on the upper canopy, the elaboration of important basic principles and many valuable suggestions.
Funding This research received no external funding.
Code availability Not applicable.

Declarations
Ethics approval This research did not involve human participants and animals. This research and its presentation in this manuscript followed the rules of good scientific practice as described on the journal website.
Consent to participate Not applicable.

Consent for publication Not applicable.
Competing interests The authors declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.