The wisheli Project - Improving the shelf-life of Umbrian white wines

Francesca Borghini and Stefano Ferrari
ISVEA srl, Via Basilicata s/n
Loc Fosci, Poggibonsi (Siena)

f.borghini@isvea.it, s.ferrari@isvea.it

INTRODUCTION

In the market of young white wines, consumer preference is for products with varietal, floral and fruity aromas as well as bright and vibrant colours. Producers make considerable efforts to meet these quality requirements from the management of the vineyard to the release of the wine. The prospect of the sensory integrity of the wine being compromised by an early flattening of the bouquet, with the consequent homogenisation of organoleptic characteristics, the appearance of off-flavours or darkening of colour, is an extremely worrying issue for producers. The possible causes of this scenario can be many: among the main factors that affect the durability of wine, in fact, are the identity predisposition of the grape varieties used, the pedoclimatic conditions and the agronomic practices adopted, but also the management of the processing, maturing and storage of wines before and after bottling until their distribution and final consumption. Given the significant negative impacts - in terms of image, competitiveness and added value - that the above-mentioned phenomena have on the wine industry, it has become inevitable for the world of wine production to increase its efforts to identify and prevent the causes of early wine quality decline.

The Wisheli project (development of new production techniques to improve the shelf-life of Umbrian wines), an innovation project for the 2014-2020 Rural Development Programme for Umbria - Measure 16 - Submeasure 16.1, has set out to make a significant contribution to the general improvement of the quality of wine production in the Region of Umbria, taking on board the problematics expressed by producers regarding a reduction in the shelf-life of young white wines

Operationally, the project was developed through the following pathway (see also Figure 1):

• Study and development of a test to assess the susceptibility of young white wines to deterioration: samples of bottled wines ready for sale were first subjected to treatment in climatic chambers, using different temperature-time combinations, and then analysed in depth from both a chemical and sensorial point of view. All the data obtained was then processed by means of multivariate statistical techniques that made it possible to identify the best time-temperature conditions and also the most significant parameters for the development of the test to assess conservation over time.
   
• A scientific investigation dedicated to studying the evolution of bottled wine, exploring the possibility of identifying representative markers of premature ageing; to this end, all the wines included in the project were stored for over a year at both an ideal temperature (13°C) and at room temperature (23°C), and analysed at regular intervals.
   
• Evaluation of the impact of certain wine production strategies on the resistance to deterioration of young wines with the aim of improving the wines' ability to maintain its quality attributes as unaltered as possible. In this phase, microvinifications were carried out on Trebbiano and Grechetto grapes, considering the impact of different variables such as vineyard management, cold maceration of the grapes and the use of a demetallizer. This was done with the aim of gathering further, valuable information, which could also be used to draw up specific operational protocols for the sector, capable of making a substantial contribution to the goal of minimising the negative impact that the factors examined, both intrinsic and extrinsic, have on the early qualitative decline of young wines.

Figure 1. Structure of the Wisheli Project (MV, microvinifications).

RESULTS

Table 1 describes the blends of bottled wines used in the Wisheli project to develop the test for assessing the propensity of white wines to age.

Table 1. Bottled wines subjected to different treatments (temperature/time) in a climate chamber.

Redundancy analysis (RDA) was carried out both to identify marker molecules of thermal stress in white wines, to identify the time/temperature conditions to be used in the predictive tests, and to select the explanatory variables of the phenomena involved.

The grape variety was used as a conditioning variable in the partial RDA, since an initial PCA analysis had shown a clustering of samples according to wine type rather than heat treatment.

The constrained inertia (i.e. the variance explained by the predictors, the heat treatments) is equivalent to 60% and the analysis was statistically significant (p<0.01). The samples are grouped according to temperature: those treated at 45°C are in the quadrant characterised by positive values of F1 and F2 and the variables that most determine this distribution and therefore indicate thermal stress, circled in red in figure 2, are 1,6-trimethyl-1,2-dihydronaphthalene (TDN), vitispirane (VTP), furfural (Fu), nicotidamide (nic) and epicatechin sulphonate (ECS). The concentrations of these molecules, both aromatic and non-aromatic, increase with both temperature and exposure time at 45°C. We also notice significant changes in the colour of the samples, as indicated by the CIELab coordinates (L*, H* and a*) and a decrease in certain esters (phenylethylacetate, isoamyl octanoate, isoamylacetate) which give the wine sweet and fruity notes.

The natural evolution (at 13°C, along the second axis, F2) is characterised by a decrease in certain polyphenols (e.g. glycosylated flavonols and caftaric acid, which presumably undergo hydrolysis) and aromatic compounds such as esters (isoamylacetate) and terpenes. Over time, there is also an increase in certain esters that give wines negative aromatic notes (ethyldecanoate → butter), diethylsuccinate (a well-known marker of ageing), and certain sulphonates of tryptophan catabolites (tryptophol and indol-lactic acid).

The samples treated at 35°C for 5, 10 and 15 days are in an intermediate position between those at 13°C and 45°C, with treatment 35-5 (35°C for 5 days) being closest to 13-90 (13°C for 90 days), which therefore appears to be the one that best simulates the natural evolution of white wines and which does not present significant differences compared to treatment 35-10 (35°C for 10 days).

To sum up, some markers of thermal stress have been identified among volatile and non-volatile molecules and, among the conditions tested for white wines, the one that best simulates natural ageing seems to be treatment 35-5.

Figure 2. RDA of the chemical parameters (left) in the samples subjected to climate chamber treatment (in red, right); in the tables the variables that weigh most on the axes (F1 and F2), F1: TDN = trimethyldiyronaphthalene, PaA = phenylacetic acid, PeA = phenylethylacetate, iAO = isoamyl octanoate; iPA = isoamylacetate; ECS = epicatechin sulfonate, nic = nicotidamide, L*,H*, a* = CIELab coordinates, CFT = kaftaric acid, d1 = 420nm absorbance; F2: Qglc = glycosylated quercetin, GRP = grape reaction product, TS = tryptophol sulfonate, DeS = diethylsuccinate, ED = ethyldecanoate, ILS = indol-lactic acid sulfonate.

Colour characteristics, polyphenol composition and aromatic substances have been identified as variables that give an indication of a white wine's propensity to age. Since the aromatic component and its variations over time are strictly variety-dependent, in order to define an index that is as practical, simple and above all independent of the blend of the samples as possible, the respective chromatic variation was studied in detail. In particular, a ΔE parameter was defined, derived from the variations in brightness (L*) and in the red/green (a*) and yellow/blue (b*) components, before and after the test treatment (35°C-5 days) in a climatic chamber. Based on the measured values, an ageing risk index was characterised (Table 2).

Table 2. Aging risk index for white wines as a function of ΔE after climate chamber test.

It is very interesting to note that the ΔE of the samples treated in the climate chamber correlates positively not only with that measured in the samples stored at 13°C for 90 days, but also with the ΔE calculated after one year's storage of the samples at 13°C and 23°C (Figure 3), confirming its validity as an indicator of the natural ageing of wines.

Figure 3. Correlations between ΔE of samples treated in the climate chamber (35-5, on the abscissae) with those calculated after one year of storage at 13° and 23°C. 

The analysis of correlations between ΔE 35-5 and the chemical parameters analysed showed statistically significant correlations (p<0.01) between ΔE and certain phenolic compounds (gallic acid, caffeic acid and flavanols), which are risk factors for sample deterioration over time.

The 5 commercial samples were analysed after one year of storage at 13°C and 23°C: Figure 4 shows the distribution of samples treated in the climate chamber and after one year. Despite a clustering of the samples according to variety, it can be seen that the wines after one year, especially those stored at 23°C, are significantly different according to the chemical variables analysed. The wine that changes the most over time is Sauvignon Blanc (which cannot be explained by the PCA model after one year). This sample presents high concentrations of iron, riboflavin and polyphenols in general and flavanols in particular. High levels of these parameters may therefore constitute risk factors for the deterioration of white wines over time.

Figure 4. PCA score plot of commercial samples subjected to climate chamber treatments and analysed after one year of storage at 13° and 23°C.
 

Finally, the climate chamber test and risk index based on ΔE was applied to the samples produced during the microvinifications foreseen by the project in 2018 and 2019 and aimed at assessing the effect of pre-fermentative maceration, the use of a demetallizer and also the quantity of grapes per hectare (high/low production) on the resistance to spoilage over time of the final product (Figure 5). In the 2018 samples, untreated trebbiano (T1) had a low risk of ageing; this risk increased slightly in demetallised (T3) and significantly in macerated and demetallised samples (T4). Grechettos already before undergoing maceration and/or demetallisation have a higher risk of propensity to decay and pre-fermentative maceration further worsens the situation (G8). In 2019, the results are completely different: grechetti, apart from the longer or shorter maceration time, are always at low risk according to the ΔE; non-macerated trebbiano, both from high and low production, are found to have, like grechetti, a ΔE value 35-5 lower than 1, therefore at low risk of ageing, while trebbiano samples subjected to a 4-hour maceration, especially those from low production, have ΔE values corresponding to medium or high risk. After two years of storage at 23°C, the samples were tested again and the index values were all above 4, which probably reflects the general deterioration of these young white wines.

Figure 5. ΔE 35-5 for the 2018 (top) and 2019 (bottom) microvinifications. At the bottom, the green circles show the ΔE of the samples calculated after 2 years of storage at 23°C. T, trebbiano and G grechetto.

Materials and methods are available in an annex at the end of the article.

CONCLUSIONS

Over the three years of the project, the analysis of wines subjected to different treatments, in terms of time and temperature in the climate chamber, has made it possible to develop a test and an index to assess the propensity of young white wines to deteriorate. The test for white wines consists of treating the wines at 35°C for 5 days and then assessing the colour change by calculating the ΔE, based on CIELab coordinates. Risk factors that could increase the propensity to age in white wines were also identified: high levels of Vitamin B2, Flavanols and Fe.

The climatic chamber test and the index were applied to samples of Trebbiano and Grechetto microvinified in 2018 and 2019 and the results are not variety dependent: in Trebbiano the pre-fermentative maceration, which enriches the samples with some aromatic compounds such as esters and terpenes (data not shown), seems to affect the longevity of the samples (both in 2018 and 2019), while the results for Grechetti are contrasting in the two microvinifications but the enrichment in aromatic compounds is not significant. The cost-benefit assessment in terms of the risk of spoilage of the oenological treatments to be applied during vinification still remains with the technician.