Investigation of an early harvest regime and pre-fermentative blending treatments to produce lower alcohol wines: impact on grape and wine composition and quality

Abstract

Warmer and shorter grape ripening periods, as well as concomitant extreme weather events like heatwaves, have been posing considerable challenges for wine producers, particularly when winemakers seek to extend fruit hang-times to meet consumer demands of fuller flavoured wines. Consequently, grapes may not be harvested at desired qualities and may be exposed to over-ripeness or berry shrivel, which likely translate into excessive wine alcohol concentrations. As the nature of these weather events is rather unpredictable and succumbs to annual fluctuations, winemakers rely on flexible and economic strategies (e.g. alternative to current physical dealcoholisation techniques) to ameliorate situations of excessive grape ripeness. Among the various methods to manage wine alcohol levels, one recently proposed strategy is the pre-fermentative dilution of sugar in juice with either a very low alcohol wine (~5 % alcohol by volume) or water. However, the effect of such manipulations on wine compositional and sensory qualities was not entirely understood. Further, it has been unclear how the resulting wines would compare to those of similar alcohol levels made from earlier harvested grapes that are picked to avoid grape over-maturity. For this purpose, studies were undertaken to evaluate these approaches to alcohol management for the production of Cabernet Sauvignon and Shiraz wines under a variety of vintage conditions. The studies have been drafted as manuscripts that have been prepared for publication or have already been submitted or published in peer-reviewed journals. The manuscripts are presented in chapters as outlined below after an introductory chapter. The first study reports on the initial vintage (2015), in which extremely warm and dry conditions caused an exemplary over-maturity of the Cabernet Sauvignon grapes. Experimental wines of various lower alcohol concentrations were produced via prefermentative substitution of juice with a very low alcohol green harvest wine (GHW) or ii water at various rates. The consequences for non-volatile wine composition (colour parameters, tannin and polysaccharide composition) of these pre-fermentative approaches were reported relative to the high alcohol control wine. The characteristics of the substituted wines were discussed in the context of wines of similar alcohol content produced from earlier harvested grapes, thereby providing an evaluation of the role of harvest decision in this extreme vintage scenario. It was shown that colour and tannin parameters were not significantly affected even by the highest substitution rates (with GHW and water) compared to the control and in fact retained values superior to those in wines resulting from earlier harvests. A manuscript detailing this work has been published in Food Chemistry, 244 (2018) 50-59. Further building on the 2015 experimental winemaking, a second manuscript presents the consequences of the alcohol management treatments on wine volatile composition (qualitative and quantitative data obtained with GC-MS analysis) and wine sensory profiles (determined with descriptive sensory analysis). Analysing the same wines as before, the substitution treatments were contrasted with wines arising from earlier harvests to outline potential merits and disadvantages of each approach, this time in terms of volatile and sensory profiles. According to the GC-MS data, the implementation of water had the least effect on the volatile composition, causing rather minor concentration changes of a small fraction of the analysed volatiles. This was further mirrored in the sensory profiles of the lower alcohol wines, which were found to be strongly reminiscent of the overripe control wine, hence there not only positive results (aroma intensity, red fruit, dark fruit) but also negative attributes (hotness) sustained when adjusting the wine alcohol levels via the proposed pre-fermentative treatments. This indicated that wine styles are more affected by harvest date than the substitution treatments. This study has been published in Food Chemistry 259 (2018) 196-206. iii The two studies reporting on the 2015 winemaking trial provided evidence that prefermentative additions of GHW or water are suitable for the production of lower alcohol wines from highly mature Cabernet Sauvignon grapes without greatly affecting the wine quality. The objective of the subsequent study in 2016 was to confirm these findings and to further evaluate an early harvest regime and pre-fermentative substitution treatments as means to produce lower alcohol wines under milder vintage conditions and with relevance to changes in regulation allowing water addition under certain circumstances. Cabernet Sauvignon and Shiraz were investigated and the resulting wines were examined for colour, tannin parameters, volatile composition and sensory properties. The benign nature of the substitution treatments on wine quality parameters, for instance stable levels of anthocyanin and tannin concentrations, was confirmed for Cabernet Sauvignon, but less so in case of Shiraz, where more pronounced differences emerged according to the blending component used. In this case, water substitution was identified as the more suitable treatment to manage wine alcohol levels under mild vintage conditions while preserving the wine quality as defined by harvest date. Wine volatile profiles were generally more affected by the blending treatments in the 2016 vintage context and as a function of the variety. Different responses for Cabernet Sauvignon and Shiraz were associated with the different blending components, however without largely influencing the volatile profiles in comparison to the controls. This study was prepared as a manuscript for submission to a peer-reviewed journal. Following a recent change in regulation that allows the addition of water into the winemaking process (and consequently lowering the wine alcohol concentrations), there has been a particular interest by the wine industry to understand the consequences on wine composition, wine style and quality to facilitate decision making around this newly available winemaking technique. In this context, the final study of experimental Shiraz wines from 2017 extended upon the conclusions drawn in the preceding studies and addressed additional gaps in knowledge about adding water during winemaking. The experiments focused on iv evaluating two options for pre-fermentative water addition during the winemaking process – that is, substitution versus addition without juice removal (i.e., dilution). In addition, the importance of grape maturity on producing high quality, lower alcohol wines with prefermentative water addition was assessed using grapes harvested at two distinct maturity levels, targeting “fresh” and “mature” stages of fruit development. Based on a lower grape maturity (i.e., Fresh Fruit), low juice substitution with water did not change colour properties, whereas the analogous dilution treatment with water elicited declines in colour intensity and stability in line with decreases in total phenolics and tannin concentrations. The juice dilution further resulted in declines of important sensory attributes, such as ‘flavour intensity’ and ‘body’, diverging from the more benign substitution treatments. When applied at a greater grape maturity level (i.e., Mature Fruit), substitution or dilution with water appeared to have a greater effect on colour properties with only small implementation volumes, but high dilution rates in particular resulted in dramatically decreased tannin concentrations. In terms of wine sensory profiles, the differences between substitution and dilution treatments appeared to be less pronounced for Mature Fruit Shiraz wines compared to the Fresh Fruit counterparts, but high implementation rates well beyond the legal limit of must dilution (minimum of 13.5 °Bé) led a noticeable decline in an array of sensory attributes. Analysis of the volatile data is underway and the study will be reported in form of a manuscript for submission to peer-reviewed journal. In conclusion, this work has provided knowledge on the consequences for wine quality associated with pre-fermentative alcohol management approaches involving the implementation of water or a very low alcohol wine into the must. The managed wines were hereby around 1% - 6% lower in alcohol by volume, so generally exceeding the capabilities of other viticultural or microbiologic strategies, but lying within the possibilities of postfermentative physical dealcoholisation technologies. Although observed implications on final wine sensory attributes were marginal particularly at low to moderate levels of alcohol v adjustment, this study has illustrated that higher grape maturities with subsequent alcohol management provides only limited merits to the wine quality, so that an earlier harvest could be a more appropriate solution. Given that this project included three distinct vintage situations and two important red wine varieties, the results can help support winemakers to make informed decisions regarding wine alcohol management according to harvest situation and preferred wine style.