Abstract
The influence of soil salinity on the chemical composition, volatile aromatic compounds, and sensory characteristics of Nero d’Avola wine was examined. Data on vineyard productivity, such as grape yield, are also reported. Physicochemical parameters were determined on the samples analyzed. Solid-phase microextraction was used for the extraction of aromatic volatile compounds, followed by capillary gas chromatography-mass spectrometry. Sensory analyses were performed by visual inspection, smelling, and tasting. Statistical analysis showed that most of the physicochemical parameters and volatile constituents, both primary and fermentation aromas, were influenced by the soil salinity. The composition differences observed among the samples had little influence on the sensory characteristics; the increase in soil salinity enhanced color intensity, purple reflexes, salty, citrus, and fruit in the aroma. Results indicate that Nero d’Avola vine may be well suited to increased soil salinity, even given reduced plant productivity.
The progressive salinization of land, a consequence of climate change, is a severe problem in agriculture production (Sidari et al. 2008, Greenway and Munns 1980, Pandey and Thakares 1997). Salts are a common and necessary component of soil and essential plant nutrients, but at high content they may affect plant life. A saline soil (ECe > 4 dS/m) inhibits plant growth by limiting water uptake through an osmotic or water-deficit effect and by limiting ion uptake, injuring cells through ionic stress effects (Munns 2002, 2005). These factors are manifested in plants through morphological, physiological, and metabolic modifications such as a decrease in seed germination, shoot, and root length (Arshi et al. 2002), alterations to the integrity of cell membranes, and the inhibition of various enzymatic activities and photosynthesis (Sairam and Tyagi 2004).
In Sicily, the total area of saline soils is ~600,000 ha, mainly concentrated in the southern and western part of the island. Vitis vinifera L., one of the primary fruit crops in Sicily in both production and economic importance, is considered moderately sensitive to salinity stress. The optimal productivity of the vine is correlated with a soil salinity that does not exceed 1.5 dS/m (Maas and Hoffman 1977, Hawker and Walker 1978, Shani et al. 1993, Walker et al. 2002); excessive salinity causes growth inhibition and CO2 assimilation due to changes in stomatal conductance, electron transport rate, leaf water potential, chlorophyll fluorescence, osmotic potential, and leaf ion concentrations (Walker et al. 1981, 2004, Schultz and Matthews 1988, Patakas and Noitsakis 1999, Ben-Asher et al. 2006, Medrano et al. 2002).
Given that, to the best of our knowledge, the scientific literature only refers to physiological and metabolic effects, this study analyzes the influence of soil salinity on the chemical composition, volatile aroma compounds, and sensory characteristics of Nero d’Avola wine. Nero d’Avola is a typical Sicilian wine, ruby red in color with purplish reflexes. It is mainly produced as varietal wine but also blended with other varieties such as Cabernet Sauvignon or Syrah. Its name refers to a village in southeastern Sicily, where the variety evolved through selection by vine growers and from where it has spread throughout the island. The analyzed wine samples were obtained from grapes grown on soils of different salinity. Solid-phase microextraction (SPME) was used to extract aroma volatile compounds and has successfully been used for the characterization of wines (López et al. 2002, Maarse 1991, Mestres et al. 2002), followed by capillary gas chromatography-mass spectrometry (GC–MS) (Verzera et al. 2008). Sensory analyses were carried out through visual inspection, smelling, and tasting. Quantitative volatile data were correlated with the results of the sensory analyses. This research is a part of a larger project aimed at enhancing the production of Sicilian wines that are suitable for pedoclimatic changes.
Materials and Methods
Sampling.
Samples of analyzed Nero d’Avola wines were produced in a vineyard located in Santa Margherita Belice (Agrigento, Sicily, Italy) at 280 m asl. The climate of the area is Mediterranean, with a dry period from May to September, a rainy period concentrated in the winter months, an average annual temperature of 17.4°C, and an average annual rainfall of 648 mm. The gently sloping vineyard faces southeast. The salt content of the soil increases along the rows from top to bottom; consequently, the vineyard was divided into three zones of different salinity: zone 1: negligible salinity, ECe 0.7 dS/m (average value up to a depth of 105 cm); zone 2: average salinity, ECe 1.2 dS/m (average value up to a depth of 55 cm) and 2.1 dS/m (average value from 55 to up to a depth of 105 cm); and zone 3: high salinity, ECe 1.0 dS/m (average value up to a depth of 55 cm) and 7.6 dS/m (average value from 55 to up to a depth of 105 cm).
The soil was defined as Sodic, Vertical, and Calcixerepts according to USDA Soil Taxonomy, with 55 to 60% clay at pH 8.1 to 8.3. The most common cations and anions are potassium, calcium, magnesium, chlorides, sulfates, and carbonates, with sodium, magnesium, and sulfates prevailing. The salt content gradually increases with depth and the maximum increase (cations = 19.39 meq/L; anions = 19.11 meq/L) was at 70 to 110 cm soil depth in zone 3. In summer, irrigation using rainwater is at times necessary to avoid an increase in salt concentration because of greater evaporation of soil water.
The grapes from the three zones were harvested separately in September 2007 and 2008 and immediately transferred to an experimental winery of the Istituto Regionale della Vite e del Vino in Marsala (Sicily, Italy), then pressed. The musts were sulfited (0.05 g/L) and dry yeast NDA21 (0.3 g/L) was added. Alcoholic fermentation occurred for 9 days at 26 to 28°C; malolactic fermentation was carried out with the addition of biomass (0.005 kg/L dregs). After decanting, the wines were bottled and stored. A total of 24 sample wines were made: 12 samples each year of four samples from each zone, from two different fermentations. Each sample was analyzed in duplicate. Chemical analyses were carried out on the grape must and, after fermentation, on the wine before bottling. Six months after bottling, the wines were analyzed for aroma compounds and to define their sensory characteristics.
Chemical analysis.
Physicochemical parameters in must (sugar, titratable acidity, and pH) and wine (alcohol concentration, pH, titratable acidity, total anthocyanins, flavonoids and polyphenols, tartaric acid, lactic acid, total dry extract, sulfates, color intensity, and hue) were determined according the EEC Official Method (Regulation no. 2676/90).
Volatile extraction: HS-SPME.
A 40-mL vial was filled with 20 mL of sample. The vial was equipped with a mininert valve (Supelco, Bellefonte, PA). Extraction was performed in the vial headspace at 30°C using a commercially available fiber housed in its manual holder (Supelco). All extractions were carried out using a divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) fiber of 50/30-μm film thickness (Supelco). The liquid sample was equilibrated for 15 min and then extracted for 20 min. During the extraction, the sample was continuously stirred. After sampling, the SPME fiber was introduced onto the splitless injector of the GC–MS under the experimental conditions reported below. The fiber was kept in the injector for 3 min for thermal desorption of the analytes onto the capillary GC column. The split-splitless injector port was maintained at 260°C. No artifacts were observed after a SPME analysis of water performed as blank analysis.
Volatile analysis: GC–MS.
A Varian 3800 gas chromatograph directly interfaced with a 2000 ion trap mass spectrometer (Varian Spa, Milan, Italy) was used to analyze the volatile compounds. The conditions were as follows: injector temperature, 260°C; injection mode, splitless; capillary column, CP-Wax 52 CB, 60 m, 0.25 mm i.d., 0.25-μm film thickness (Chrompack, s.r.l. Milan, Italy); oven temperature, 45°C held for 5 min, then increased to 80°C at a rate of 10°C/min and to 240°C at 2°C/min; carrier gas, helium at a constant pressure of 10 psi; transfer line temperature, 250°C; acquisition range, 40 to 200 m/z; scan rate, 1 μs. Each component was identified using mass spectral data NIST’98 (NIST/EPA/NIH Mass Spectra Library, version 1.7), linear retention indices, literature data, and the injection of standards where available. The linear retention indices (LRI) were calculated according to Van den Dool and Kratz (1963). The repeatability of the method developed was determined by analyzing two different samples of the same wine under identical experimental conditions; the absolute peak area obtained for each component identified during three different analyses was tabulated, and the coefficient of variation (CV) was calculated. The CV was <10% for all the components identified, as previously reported (Verzera et al. 2008, Scacco et al. 2010).
Quantitative analysis.
The main compounds in the samples analyzed were quantified; each peak quantified was required to have a minimum signal to noise ratio (s/n) of 5. Quantitative results were obtained using standard additions. Standard solutions were added to multiple aliquots of each sample wine. The sample alone was also analyzed. The quantification was based on a calibration curve generated by plotting the detector response versus the amount spiked of each standard. Each sample measurement was repeated twice. The standards used were purchased from Sigma-Aldrich s.r.l. (Milan, Italy) at the highest purity available. To quantify compounds that did not have available standards, the calibration curve of a compound of the same class of substances with the most similar peak area was used (Verzera et al. 2008).
Sensory analysis.
The sensory profile (UNI 2003) was determined by a panel of 10 expert judges, three females and seven males, selected and trained according to ISO 2008. Reference standards with minor modifications were available to define descriptors (Noble et al. 1987). During training, the judges generated descriptive terms from a list of 33 attributes. A list of descriptors was selected on the basis of the frequency of occurrence of the terms used. The final set consisted of 17 descriptors: two referring to appearance (color intensity and purple reflexes), 11 referring to aroma (fruity, citrus, wild berries, fruit in the spirit, cherry in the spirit, ripened fruit, dried fruit [nut, hazelnut], floral, vegetative/herbaceous, spicy, and vanilla), and four referring to oral perception (acid, salty, bitter, and astringent). All evaluations were conducted from 10:00 to 12:00 am in individual booths (ISO 2007) illuminated with white light. The different descriptors were quantified using a 5-point intensity scale (ISO 2003). Fifty mL of each wine was served at 22°C ± 1°C (room temperature) in glasses (ISO 1977) labeled with a 3-digit code and covered to prevent volatile loss. The order of presentation was randomized among judges and sessions using FIZZ software (ver. 2.00M, Biosystèmes, Couternon, France). Water was provided for rinsing between wines. All data were registered on a direct computerized registration system.
Statistical analysis.
Chemical data were subjected to analysis of variance (ANOVA), Pearson’s correlation, and PCA using Statgraphics Plus software (ver. 5.1). Sensory data were subjected to ANOVA. Duncan’s multiple range test was applied to the chemical and sensory data to identify any significant differences between the samples analyzed. The model was statistically significant, with p < 0.05. PCA results were validated using the leave-one-out cross-validation method.
Results
Data regarding plant productivity in the Nero d’Avola vineyard are shown (Table 1⇓). The grape yield decreased with increasing soil salinity from 2.050 kg for plants in zone 1 to 1.304 kg in zone 3. Similarly, the average weight of the grape bunches decreased from 202 g (zone 1) to 188 g (zone 3), and the number of grape bunches per plant decreased from 10.1 (zone 1) to 6.5 (zone 3). Moreover, the increase in soil salinity from zone 1 to zone 3 reduced leaf area by ~32%.
There were no significant differences in sugars, total titratable acidity (g/L), or pH measured in the must from the three different zones. Conversely, significant differences were found in the sample wines. In particular, increases in tartaric acid, polyphenols, anthocyanins, flavonoids, and sulfates were found with increasing soil salinity; similarly, increases in color intensity and hue were also observed (Table 2⇓).
With regard to the volatile fraction, 52 components were identified in each sample analyzed: esters, fatty acids, alcohols, monoterpenes and sesquiterpenes, and aromatic compounds (Table 3⇓). Numerous esters and terpenes were identified, but the main components were ethyl octanoate (banana, fruit, fat), ethyl hexanoate (apple peel, fruit), ethyl decanoate (grape, fruity), and linalool (fresh, lavender), respectively.
Using ANOVA and Duncan’s multiple range test, statistically significant differences were found in the average concentration of most components in samples from the three different zones. Comparison of the average values for volatile compounds (Table 3⇑) revealed that most components quantified had significantly higher concentrations in zones 2 and 3. Zone 3 samples had the highest values for all identified compounds except β-phenylethyl alcohol, β-phenylethyl acetate, and the hydrocarbon sesquiterpenes, where the highest concentrations were in zone 2 samples. The total amount of esters was higher in samples from zones 2 and 3, which showed similar values; thus, the samples from zone 1 had the lowest concentrations of esters and of alcohols, acids, terpenes, and C13-norisoprenoids. Zone 2 samples had the highest volatile acids, in agreement with the highest concentrations of titratable acids and the lowest pH, and samples from zones 2 and 3 had similar amounts of alcohols, terpenes, and C13-norisoprenoids.
The correlations between chemical data and the salinity of the soil were expressed by Pearson correlation coefficients (Table 4⇓). Most of the variables, both volatile constituents and physicochemical parameters, showed a significant correlation and were thus submitted to principal component analysis (PCA). The first three principal components accounted for 89% of total variance (68.6% of total variance for PC1, 17.6% for PC2, and 2.8% for PC3) (Figure 1⇓). The compounds most strongly correlated with the first three principal components are listed (Table 5⇓). PC1, which evidenced that the wines from three zones were clearly distinct, displayed a strong correlation with most of the esters, color intensity, isoamyl alcohol, polyphenols, color hue, and anthocyanins. PC2 and PC3 separated wines from zone 2 from the others; the variables correlating most strongly with these axes were β-phenylethyl alcohol, decanoic acid, and (Z)-nerolidol for PC2 and β-pinene, hotrienol, α-muurolene, and β-damascenone for PC3.
For sensory data, the ANOVA showed no statistically significant differences among the samples for all the descriptors except color intensity, purple reflexes, fruit in the spirit, citrus, and salty, which were highest in zone 3 (Table 6⇓).
Discussion
The results demonstrated that soil salinity affected plant productivity (Table 1⇑), in agreement with Walker (1994), who reviewed the response of vines to salinity. Data from the literature show that the effects of salinity on vine performance are related to shoot growth and yield responses to root growth. (Prior et al. 1992a,b,c, Walker et al. 1996, Ben-Asher et al 2006). Consistent with the soil salinity, the descriptor “salty” was highest in zone 3, medium in zone 2, and lowest in zone 1. The Pearson’s correlation showed that most of the physicochemical parameters and volatile constituents, both primary and fermentation aromas, were influenced by the soil salinity and the principal component analysis showed a clear distinction between wines from soils with different salinity. Esters, color intensity, isoamyl alcohol, polyphenols, color hue, and anthocyanins were the main contributors to the first component. Volatile esters are responsible for the fruity character of fermented beverages and thus constitute a vital group of aromatic compounds in wine. In our samples, the different concentrations of esters in the wines from the three zones could be due to different precursor (free fatty acids) availability. It has been demonstrated that provision of medium-chain fatty acids to the fermentation medium causes a strong increase in the formation of the corresponding esters, thus supporting the hypothesis that precursor availability has an important role in ester production (Saerens et al. 2008). From our data, soil salinity had a positive effect on ester concentration, which increased from zone 1 to 3, as occurred for polyphenols and anthocyanins, in agreement with previous studies on other red grape varieties (Hardie and Considine 1976, Matthews and Anderson 1988, Koundouras et al. 1999). The different concentration of anthocyanins among the three zones can be correlated with the visual sensory descriptors and the chemical data, such as color intensity and hue, while that of esters correlates with the aroma descriptor “fruit in the spirit.” Among fermentation aromas, isoamyl alcohol and β-phenylethyl were the strongest contributors to the first and second component, respectively. These volatile constituents can be related to the presence of free amino acids in the must. In fact, β-phenylethyl alcohol is derived from phenylpyruvate, the direct precursor of phenylalanine, while isoamyl alcohol derives from leucines (Clarke and Bakker 2004). The concentrations and proportions of free amino acids, which break down during fermentation, are responsible for some differences in volatile compound concentration. Thus, it would seem that the effect of soil salinity on grape composition also concerns amino acid concentration.
Among the primary aromas, the monoterpenes, sesquiterpene hydrocarbons, and C13-norisoprenoids were the main contributors to the third component. To the best of our knowledge, the literature has no information on terpene constituents of Nero d’Avola grape or wine, only on grape pomace and stalks (Ruberto et al. 2008). Terpenes have a pleasant aroma and a very low olfactory threshold and are therefore perceived during winetasting even in low concentrations. Due to several synergic and antagonist effects, they correlated with the citrus descriptor. They are mainly derived from the grape, synthesized during maturation, and qualitatively and quantitatively influenced by cultivar, soil, climate, and viticulture practices. These compounds are not greatly modified during the fermentation processes, so that their presence in wine is directly related to their presence in must and therefore in the grapes (Pena et al. 2005). The fruit-derived C13-norisoprenoids, such as β-damascenone (sweet and apple) and geranyl acetone (fresh rose floral), are important odorants in wines and are thought to originate from carotenoid degradation.
Conclusion
Agronomic, sensory, and chemical results show the influence of soil salinity on the quality of Nero d’Avola wine. Moreover, the data can be used to characterize this red Sicilian wine. Statistical analysis of the data makes it possible to confirm that soil salinity influenced wine composition, and thus the whole grape composition. The amounts of polyphenols, anthocyanins, primary aroma compounds, and C13-isoprenoids, which are derived from the grapes, were significantly different with increasing soil salinity. Moreover, fermentation aromas, such as esters, β-phenylethyl alcohol, and isoamyl alcohol, were also influenced by soil salinity; the different amounts of these volatile constituents can be due to precursor availability in the musts, such as amino acids and fatty acids. The compositive differences observed among the samples had little influence on the sensory characteristics. Interestingly, the increase in soil salinity enhanced color intensity, purple reflexes, salty, citrus, and fruit in the spirit aroma. Moreover, the wines from medium and high saline soils were preferred by wine experts while those from negligible salinity were defined as flat and dull.
Nero d’Avola vine may be considered well suited to increased soil salinity even with reduced plant productivity, which is important information given the need to augment the cultivation of vine varieties that are resistant to climate change.
- Received January 2010.
- Revision received March 2010.
- Revision received May 2010.
- Accepted May 2010.
- Published online December 2010
- Copyright © 2010 by the American Society for Enology and Viticulture