On‐the‐go sensing of grape berry anthocyanins during commercial harvest: development and prospects

Background and Aims: The development and adoption of Precision Viticulture approaches to grape and wine production have been hindered by the lack of a commercially available sensor for on‐the‐go sensing of fruit quality during harvest. In this work, we sought to deploy the Multiplex®, a fluorescence‐based non‐contact hand‐held optical sensor on a harvester, for on‐the‐go sensing of berry anthocyanins during the South Australian vintage of 2010. Methods and Results: Measurements made of anthocyanin concentrations in the laboratory using the Multiplex showed high correlation (R² > 0.9) with those made on the same grapes using the standard spectrophotogrametric method. When used in hand‐held mode in the field, data collected using Multiplex demonstrated a similar spatial structure to that observed in other data layers (remotely sensed vigour, yield, elevation). Similarly, when deployed on a harvester as an on‐the‐go sensor, data obtained using Multiplex exhibited the expected spatial structure. Conclusions: Meaningful measurement of grape berry anthocyanins on‐the‐go during harvest is feasible using Multiplex. Significance of the Study: This is the first time that berry colour has been sensed on‐the‐go during harvest. The work therefore paves the way for a greater focus on attributes of fruit quality in the delineation of vineyard management zones and implementation of Precision Viticulture.     

Rapid and non‐destructive method to assess in the vineyard grape berry anthocyanins under different seasonal and water conditions*

Background and Aims: Monitoring of anthocyanins (Anth) in winegrape (Vitis vinifera L.) is fundamental for the production of top‐quality red wines. This work was aimed at testing a new fluorescence‐based sensor for Anth detection in the vineyard. Significance of the Study: The present study showed a new important innovative technology for viticulture. The sensor evaluation of the large spatial and temporal heterogeneity in Anth accumulation can be useful as support parameter in the harvest date decision or for vineyard zoning of phenolic maturity. Methods and Results: Anth in grape (cv. Aleatico) bunches attached to the vine were monitored non‐destructively in the field using a fluorescence‐based sensor during the 2008 and 2009 seasons and under different water regimes. The ANTH_RG index = log(far‐red‐fluorescence_R/far‐red‐fluorescence_G), with fluorescence signals excited with red (R) and green (G) light, was inversely correlated through an exponential function (r² = 0.875) to the Anth concentration derived from the HPLC analysis of berry skin extracts. ANTH_RG was effective in detecting the earlier ripening process in 2009 with respect to 2008 and differences in the Anth accumulation between seasons and in relation to different water regimes. Water deficit imposed in 2009 enhanced Anth concentration in berries because of a reduction in berry size but also an increase of Anth biosynthesis. This effect was observed by both destructive and ANTH_RG non‐destructive measurements. Conclusions: Our results show that the employed fluorescence sensor represents a reliable, rapid and non‐invasive tool for monitoring and determining Anth accumulation in situ.     

Suitability of pre‐dawn and stem water potential as indicators of vineyard water status in cv. Tempranillo

Background and Aims: Irrigation management requires the collection of up‐to‐date information that allow growers to make rapid decisions, water potential being one of the most used parameters in irrigation scheduling. The goal of this research was to determine whether predawn (Ψpd) and stem (Ψs) water potential can be used equally well, quantifying the effect external factors have on the relationship between them. Methods and Results: Ψpd, mid‐morning (Ψs‐m) and noon (Ψs‐n) stem water potential were measured at seven cv. Tempranillo vineyards. Climatic conditions, vine growth and fruit load affected daily water potential dynamics. A greater decline of Ψs occurred when temperatures were higher and there was no dew formation, as well as under larger canopies and fruit load. While the relative impact of these variables on water potential was not particularly high, it was by no means negligible. Conclusions: Data collected in this study did not favour Ψpd, Ψs‐m or Ψs‐n as a preferred measure to evaluate plant water status. It is important to remark that similar Ψpd values do not necessarily indicate the same level of deficit, and that similar Ψs values do not always mean similar water availability. Therefore, a straightforward interpretation of Ψpd and Ψs can sometimes be misleading. Significance of the Study: The results obtained provide significant explanation to the lack of agreement in the literature regarding the most suitable time to measure water potential. Vineyard managers and scientists should select it according to the level of stress expected, climatic conditions and irrigation frequency.     

Spatial variation in response to ‘reduced input’ spray programs for powdery mildew and botrytis identified through whole‐of‐block experimentation

Background and Aims: Previous work has highlighted the utility of the whole‐of‐block approach to viticultural experimentation. Here, we use this approach as a means of evaluating alternative spray programs for managing two economically important grapevine diseases at commercial scale. Methods and Results: Two experiments were conducted in vineyards in the Coal River and Rokeby districts of Tasmania. The first, carried out in a 4.5 ha vineyard planted to Vitis vinifera L. cv. Pinot Noir, sought to evaluate fungicide programs for control of powdery mildew (Erysiphe necator), in particular with a view to reducing the amount of sulphur applied in organic production systems. The second sought to evaluate the benefit (if any) of changing the time of fungicide application from flowering to pre‐bunch closure for the control of botrytis bunch rot (Botrytis cinerea) in a 2.4 ha Chardonnay vineyard. In both cases, treatments were applied to a whole‐of‐block experimental design using commercial spray equipment. Disease severity and the response to the various spray treatments, was spatially variable and related to topographic variation. Botrytis severity was also related to vine vigour. The powdery mildew results supported the retention of sulphur in organic spray programs, while Switch® (Syngenta Group, Basel, Switzerland) applied at pre‐bunch closure in the botrytis experiment reduced disease severity relative to the application at flowering. Spatial variation in the response to the latter treatment, when examined four times pre‐harvest, suggested that secondary spread may not have been an important mechanism for increasing disease severity over time. Conclusions: Consistent with the other work described in this series of papers, we conclude that the whole‐of‐block experimental approach offers both researchers and vineyard managers a means of acquiring more useful information than would be obtained from a conventional plot‐based approach. Significance of the Study: This study highlights the value of collecting spatially distributed data as a means of better understanding the incidence, spread, progression and control of fungal grapevine diseases. As such, it provides a new application for this experimental approach, which enables spatial variability to be used as an experimental tool.     

Whole‐of‐block experimentation for evaluating a change to canopy management intended to enhance wine quality

Background and Aims: Earlier papers have shown the utility of the whole‐of‐block approach to viticultural experimentation for assessing the relative merits of different mid‐row management strategies to improve vine vigour, and ‘reduced input’ spray programs for the control of fungal disease. Here, a similar approach was used to evaluate alternative canopy management strategies as a means of improving grape and wine quality in a Shiraz vineyard in the Coonawarra region of South Australia. Methods and Results: Two different canopy management strategies were compared: the existing practice of spur pruning to 45–50 buds per vine and shoot trimming (control) in comparison to spur pruning to 35 buds per vine, shoot thinning and shoot trimming (SPST). Contrasting treatment effects, both in terms of treatment response and its spatial extent, were observed during the trial. In 2005, the SPST treatment led to a significant improvement in some fruit quality parameters and a decline in yield. In 2006, significant treatment effects were only observed for bunch weight and the number of berries per bunch. Whilst reasons for these contrasting effects could not be deduced from the available data, it was demonstrated that the inherent variability of vine vigour within the block had a greater effect than the applied treatments. Conclusions: Consistent with the other whole‐of‐block experiments described in this series, we conclude that this experimental approach offers advantages in comparison to plot‐based approaches to viticultural experimentation. The spatial information obtained in this study created a better understanding of inherent vineyard heterogeneity and the effects this has on control options for vine vigour and grape composition than would have derived from a plot‐based experiment. Significance of the Study: This study is the first application of the whole‐of‐block experimental approach to assessment of alternative canopy management options. The work provides further demonstration of the value which accrues through using spatial variability as an experimental tool.     

Airborne high‐resolution images for grape classification: changes in correlation between technological and late maturity in a Sangiovese vineyard in Central Italy

Background and Aims: The goal of this study was to investigate the relationships between NDVI values at different phenological stages and measurements of grape parameters at two different harvest dates. Methods and Results: The research was done on a Sangiovese vineyard in Central Italy. Over four seasons, airborne NDVI measurements acquired between June and August were related to grape parameters (yield per vine, pH, °Brix, anthocyanins and polyphenols) at technological harvest (H1) and two weeks later (H2). Correlations were higher at H1 and decreased at H2 with a different rate depending on the parameter. °Brix and pH correlations showed a moderate rate of variation between H1 and H2; bigger differences and a different inter‐annual dynamic were observed in anthocyanins and polyphenols between H1 and H2. Conclusions: The ability of NDVI to discriminate different grape classes was confirmed, but its efficacy substantially varies depending on the harvest date. These results suggest the existence, within the same vineyard, of different grape populations having specific timing and shape of ripening curve; as a consequence, distinct vigour zones of the vineyard show a different evolution of the content of grape parameters between the two harvests thus influencing the degree of correlation between grape quality and NDVI measurements. Significance of the Study: This is the first study in which harvest date has been considered for its influence on the predictive skill of RS. It therefore highlights not only the importance of spatial variation within the single vineyard, but also the importance of ripening dynamics.     

Ripe rot of south‐eastern Australian wine grapes is caused by two species of Colletotrichum: C. acutatum and C. gloeosporioides with differences in infection and fungicide sensitivity

Background and Aims: Ripe rot of grapes is widespread in subtropical vineyards of south‐eastern Australia. We aimed to re‐evaluate the causative agent(s), berry susceptibility, fungicide sensitivity and differences in histopathology. Methods and Results: Both Colletotrichum acutatum and C. gloeosporioides were shown to be responsible for ripe rot in three vineyards surveyed in the Hastings Valley, NSW in 2007 and 2009. Observation on detached berries of Vitis vinifera (cv. Cabernet Sauvignon and Chardonnay) revealed that C. acutatum had a faster infection rate than C. gloeosporioides. C. acutatum also formed appressoria and penetrated grape tissue faster than C. gloeosporioides, which produced longer hyphae on the berry surface before penetration. Both C. acutatum and C. gloeosporioides produced acervuli 48 h after inoculation. In contrast, spore germination rates and growth rates on potato dextrose agar were greater for C. gloeosporioides. The two species differed in their sensitivity to the fungicides benomyl, captan and triadimenol. C. acutatum was significantly more sensitive to captan and triadimenol, whereas C. gloeosporioides was more sensitive to benomyl. Conclusions: Both C. acutatum and C. gloeosporioides cause ripe rot of wine grapes in Australia. Subtle differences in the infection process may explain the relative prevalence of the two species. Significance of the Study: This study confirms the involvement of C. acutatum and C. gloeosporioides in ripe rot of grapes in Australia. Knowledge of growth characteristics and fungicide sensitivity of ripe rot pathogens should aid disease management strategies.