Sensory and physicochemical characterization of different Mediterranean Malvasia wines Carmela Maria Lanza DOFATA, Food Technology section University of Catania, Italy e-mail:
[email protected]
Agata Mazzaglia DOFATA, Food Technology section University of Catania, Italy e-mail:
[email protected]
Ella Pagliarini DISTAM, Food Technology section University of Milan, Italy e-mail:
[email protected]
Abstract The aim of the present study was to determine the identity of different Mediterranean Malvasia wines, correlating sensory attributes with physicochemical parameters (pH, alcohol content, °Brix, titratable and volatile acidity, reducing sugars, Hunter’s colour parameters L*, a*, b* and intensity and tone of colour by spectrophotometry). The sensory profile was defined by a trained panel of 9 judges (UNI 10957, 2003). Reference standards (Noble et al., 1987) were available to train the judges in several preliminary sessions. The descriptors generated by the assessors were selected on the basis of the frequency of the terms used (%). The numerous sensory attributes of the final set, referred to appearance, smell, taste and mouth feel perceptions, were quantified using a nine-point intensity scale. The assessors were requested to evaluate the intensity of each attribute by assigning a score between 1 (absence of the perception) and 9 (maximum intensity of perception). In this study, 20 commercial samples of Malvasia wines from Lipari island, Sardinia, Crete, Istria and Canaries were analyzed in triplicate. The sensory data for each attribute were submitted to Analysis of the Variance (ANOVA) with samples (S), judges (J), replicates (R) and their respective interactions JxS, SxR and JxR, as effects. The significance of these effects was evaluated by the F test. The sensory profile allows to identify the quali-quantitative differences among the different samples of Malvasia wine underlined by Principal Component Analysis (PCA) applied to mean values of sensory and physicochemical data. Keywords: Malvasia wines, sensory profile, physicochemical parameters. Thematic Area: enology.
1. Introduction. Coming back from the IV Crusade (1202) the Venetian ship owners were attracted by the rich vineyards of the Peloponnese, in which the Malvasia cultivar was already grown. They spread these shoots in Dalmatia, Istria, up to France, Spain and Balearic islands. This is how, in the whole Mediterranean area, the diffusion of these vines started, together with the consequent differentiation of the wines owed to the different pedoclimatic and technological conditions (Calò, 2005). Among these wines, the DOC (Denomination of Controlled Origin) Malvasia delle Lipari is noteworthy, produced particularly in the island of Salina, which contributes to the variety of DOC and DOCG (Denomination of Controlled and Guaranteed Origin) wines in Sicily (Lo Iacono, 2008). This is a sweet wine obtained from the natural fermentation (Agnolucci et al., 2007) of a blend of ‘Malvasia delle Lipari’ and ‘Corinto nero’ grapes (95% and 5%, respectively), having a alcohol content of 20% vol. The grapes are generally harvested in September and exposed to sunlight on rush mats for about 10–15 days, to concentrate sugars up to about 300 g/L. Another product is DOC “Malvasia di Bosa” obtained from the vine Malvasia of Sardinia with a tolerance in the vineyards of maximum 5% of different vines. The traditional classical typology is characterized by a alcohol degree higher than 15% and maturation in barrels in presence of yeasts "flor" Saccharomyces, able to colonize the wine surface forming a biofilm (Farris et al., 2008). The wine-making industry in the Canarian islands is currently in full growth and has two peculiarities. Firstly, the climate is very different from the Mediterranean one, in which vines were first grown (Lopez et al., 2003). Secondly, the fact that the Canaries are a group of islands, has permitted the conservation of some grape varieties introduced from Europe prior to the phylloxera epidemic. Concerning these wines, many studies have been carried out, such as on postharvest drying in dehydration rooms (Costantini et al., 2006), phenolic compounds (Ramos et al., 1999) and the very interesting aromatic component in these wines obtained by the modern production techniques (Bonino et al., 2003; Camara et al., 2004; Alves et al., 2005), but few researchers have considered the sensory characteristics which determine the identity of Malvasia wines with different geographic origin. 2. Aim. In this study it is proposed a further contribution (Lanza et al., 2008) to the creation of an improved sensory profile for some Malvasia wines of the Mediterranean area, developing an appropriate and richer sensory terminology than the few sensory descriptors generally established in the DOC disciplinary (D.P.R. 20/9/1973 for the Malvasia delle Lipari), correlating them with some
2
physicochemical parameters (pH, alcohol content, °Brix, titratable and volatile acidity, reducing sugars, Hunter’s parameters L*, a*, b* and intensity and tone of colour by spectrophotometry).
3. Materials and methods. Wines Twenty commercial samples of Malvasia DOC wines: 10 from Lipari (L), Aeolian Islands, 6 from Sardinia (S), 2 from Canaries (Ca), 1 from Crete (Cr) and 1 from Istria (I) were collected and analyzed in triplicate.
Sensory analysis In the present study the sensory profile (UNI 10957, 2003) was constructed using a trained panel of 8 judges, students of the Department of Food Science and Technology, University of Catania (Italy). To train the judges and to develop a common dictionary for the generation of the descriptors, Malvasia samples were evaluated three times a week for four subsequent weeks. Reference standards (Noble et al., 1987) were available to define descriptors and every term was described and explained for removing any doubt. A list of descriptors was selected on the basis of occurrence (%) of the terms generated, and references for every attribute, corresponding to the maximum intensity of the nine point intensity scale chosen, were used (Lanza et al., 2008). The final set consisted of twenty four attributes: two referring to appearance (colour, clarity) twelve referring to aroma (citrus, alcohol, apricot, caramel, floral, fruity, exotic fruit, stewed fruit, woody, yeasty, spicy, raisins), three to taste (acid, sweet, salt), five to flavour (apricot, fruity, honey, wood, raisins), astringent and burn (referring to mouth feel perception). Each judge evaluated three wines in every session using a nine-point scale to quantify the different descriptors. The evaluations were carried out from 10.00 to 12.00 a.m. in individual booths illuminated with white light. Thirty mL of each wine was served at 22°C ± 1°C (room temperature) in glasses labeled with a 3-digit code and covered to prevent loss of volatiles. The order of presentation was randomized for each judge and in each session. Water was provided for rinsing between wines.
Physicochemical analyses Determination of pH and titratable acidity was performed according to the AOAC methods (1990). Colour measurement was carried out in 1-mm path length coupling cuvette using a Varian Cary 1E spectrophotometer (Varian Inc., Melbourne, Australia) equipped with Cary Color Calculation 3
software to record the L*, a*, b* values of the samples from the Hunter system and the absorbance at 420 (yellow), 520 (red) and 620 nm (violet), to calculate the intensity (I = λ420+λ520+λ620) and tone (N = λ420/λ520) of the colour.
4. Results and discussion. All chemical analyses were carried out in triplicate and the data obtained were reported as means with standard deviations. The ANOVA elaboration for the physicochemical data was carried out separately on samples from Lipari and Sardinia (data not shown). Results showed significant differences for every considered physicochemical parameter, at p ≤ 0.001 between the Lipari and Sardinia groups. Samples from Istria, Crete and Canarian islands were not subjected to ANOVA elaboration, considering the low number of samples. Results for the physicochemical analyses are reported in table 1. The sensory data for each attribute were submitted to analysis of variance (three-way ANOVA), with samples, judges, replicates and their interactions as effects. The significance of these effects was assayed by the F-tests. The sample means averaged across judges and replicates were submitted to Principal Component Analysis (PCA) in order to interpret differences amongst wine samples using Unscrambler® 8.0 software (Camo As, Trondheim, Norway). Each sensory attributes shows significant differences for all descriptors at p≤ 0.001, except for the spicy aroma descriptor. F values for replicates and interaction Judges×Replicates and Samples×Replicates are not significant for nearly all the attributes . These results indicate that the mean scores given by the panel for each attribute of Malvasia wines, are satisfactory estimates of their sensory profiles. The sensory data means of Mediterranean Malvasia wines of the 5 zone are been using to construct the sensory profile of them reported in figure 1. As a result, they were averaged across assessors and submitted to Principal Component Analysis (PCA) with instrumental data. A multidimensional space (Bi-blot) based on sensory and physicochemical data was obtained (Figure 2). The variance explained by the first two principal components was 62%. The Bi-plot shows that the Malvasia wine samples are well separated in the space. Moving along the first component (PC1) (explained variance 43%), Istria (I1, I2 and I3) wines are grouped on the left while Crete (Cr1, Cr2 and Cr3), and five Lipari Malvasia (L3, L5, L6, L 9 and L10) wines are on the right.
4
Istria samples are not characterized by any sensory attribute except the aroma of yeasty, astringent and sour. Crete and Lipari samples are characterized by the aroma attributes of spicy, honey, apricot, stewed fruit, fruity, flavour of spicy and burn, as a mouth feel descriptor. The second component (PC2) (explained variance 19%) distinguishes the Sardinia samples (S2, S4 and S6) from Lipari (L2, L4, L7 and L8) and Canary (Ca1 and Ca2) ones. The samples from Sardinia in the upper part of the graph are separated from the other and characterized by the aroma attributes of woody, citrus and alcohol, by the flavour of woody and by the colour parameters a*, b*, Intensity, and by the alcohol content. These four Lipari samples on the bottom right of the plot, are distinguished from the rest of the Lipari products and they are characterized by the aroma of exotic fruit and floral, by the flavour of fruity, honey and raisin and by the physicochemical parameters pH, volatile acidity, °Brix and reducing sugars. The few Canarian wine samples on the bottom left of the plot are negatively correlated only with sensory attributes, while are positively correlated with titratable acidity and with the colour parameter L*.
5. Conclusions Among the others, Malvasia delle Lipari represents a niche product which is obtained in a limited area with a special technique of sun drying, which allows the wine to gain and concentrate the most peculiar traits of the volcanic environment, and to be characterized, apart from sweetness which is correlated with the °Brix and reducing sugars, by a wide and harmonic aromatic profile where the aroma of floral, fruity and exotic fruit, flavours of honey, fruity and raisin, prevail. The similarity of sensory characteristics between Lipari and Crete samples, the latter showing honey, apricot, raisin and stewed fruit aroma, could origin from the geographic proximity and from the frequent cultural exchanges between southern Italy and Greece. Malvasias di Bosa (Sardinia) can boast their own identity, based on prevailing aroma of citrus and wood, flavour of wood, high alcohol content, and distinctive colour parameters, consequence of the characteristics of the territory and of the scarce contacts with other areas. Canarian Malvasias, very far, geographically, from Lipari ones, represent a peculiar entity for the specificity of territory and production technologies, which result in wines characterized by high clarity and acidity, without particular aromatic attributes. Istrian Malvasias, among the most ancient, being derived from the first Malvasia vines settlings, astringent and with an aroma of yeast, show a profile which cannot be compared to any of the other samples.
5
From the results of the present research it can be inferred that Malvasia cultivars distributed in the Mediterranean basin for more than 500 years, have been modelled by the pedoclimatic environments and by the growing techniques, connected to multiple cultural factors, that each population has developed with time. Such interaction has leaded to products with sensory characteristics which are qualitatively and quantitatively different from each other, contributing to expand the range of Malvasia wines nowadays available.
6
Table 1. Results of physicochemical analyses.
I
3.32±0.03
Titratable acidity (g/L tartaric ac.) 0.62±0.01
Cr
3.65±0.01
0.59±0.01
1.05±0.01
18.25±0.00
25.75±0.52
16.22±0.19
90.19±0.02
-0.65±0.03
33.13±0.03
0.82±0.00
3.56±0.01
S1
3.30±0.05
0.59±0.01
0.57±0.01
11.23±0.06
23.41±0.00
16.26±0.33
96.50±0.16
-3.15±0.05
15.45±0.05
0.31±0.00
4.71±0.02
S2
3.26±0.01
0.59±0.01
0.53±0.00
11.00±0.00
10.51±0.30
17.84±0.61
82.94±0.23
2.97±0.12
49.73±0.29
1.30±0.01
2.75±0.01
S3
3.28±0.01
0.58±0.00
0.54±0.01
9.00±0.00
3.08±0.03
17.28±0.05
74.98±0.07
2.53±0.00
38.17±0.15
1.45±0.00
2.06±0.00
S4
3.52±0.02
0.52±0.01
0.86±0.02
10.30±0.00
3.87±0.06
17.42±0.19
91.90±0.12
-2.96±0.13
26.45±0.04
0.66±0.00
3.28±0.00
S5
3.74±0.01
0.50±0.00
1.10±0.02
10.60±0.00
6.48±0.04
18.12±0.14
65.31±0.70
3.81±0.03
38.69±0.37
2.01±0.01
1.96±0.01
S6
3.42±0.07
0.53±0.01
0.73±0.01
9.00±0.00
2.61±0.03
14.98±0.60
90.76±0.12
-1.33±0.06
34.07±0.04
0.69±0.00
2.97±0.00
L1
3.46±0.04
0.56±0.00
0.69±0.03
13.90±0.10
41.87±1.45
11.73±0.00
96.84±0.20
-4.06±0.03
19.90±0.01
0.38±0.00
5.41±0.02
L2
3.45±0.06
0.49±0.01
0.61±0.04
21.23±0.06
88.03±1.06
14.65±0.00
90.61±0.03
-0.18±0.02
37.11±0.04
0.97±0.01
2.91±0.03
L3
3.63±0.06
0.57±0.01
0.94±0.05
17.40±0.00
54.79±0.00
13.3±0.00
94.53±0.08
3.49±0.03
23.916±0.06
0.55±0.01
4.08±0.04
L4
3.40±0.05
0.61±0.00
0.48±0.02
21.37±0.06
91.96±0.00
19.84±0.09
94.50±0.02
-3.14±0.02
24.69±0.07
0.52±0.01
3.61±0.05
L5
4.10±0.06
0.58±0.01
1.28±0.03
20.03±0.06
65.20±1.17
14.29±0.00
92.83±0.04
3.96±0.02
32.08±0.01
0.75±0.00
4.22±0.01
L6
3.81±0.01
0.49±0.01
0.98±0.07
15.00±0.00
20.52±0.57
15.66±0.37
96.21±0.10
-4.54±0.04
21.48±0.00
0.44±0.01
5.53±0.33
L7
3.96±0.01
0.61±0.01
1.35±0.07
24.00±0.00
93.64±3.41
15.25±0.42
85.13±0.17
0.19±0.08
49.14±0.14
1.38±0.00
3.86±0.02
L8
3.39±0.02
0.60±0.02
0.93±0.04
17.50±0.00
33.44±0.43
10.36±1.72
96.40±0.00
-2.55±0.16
15.8±0.03
0.34±0.00
4.35±0.11
L9
3.77±0.01
0.56±0.02
0.79±0.02
20.90±0.00
51.69±2.53
14.00±0.00
83.44±0.19
0.05±0.02
49.16±0.38
1.49±0.00
3.66±0.01
L 10
3.81±0.00
0.49±0.01
0.98±0.06
15.00±0.00
20.53±0.47
15.66±0.30
96.21±0.08
-4.54±0.03
21.48±0.00
1.38±0.00
3.86±0.02
Ca 1
3.30±0.01
0.73±0.01
0.70±0.00
19.30±0.00
69.65±1.54
12.50±0.00
98.41±0.35
-1.17±0.20
5.34±0.33
0.11±0.01
4.29±0.30
Ca 2
3.34±0.05
0.68±0.02
0.50±0.02
7.83±0.17
8.31±0.00
12.30±0.00
99.61±0.96
-1.52±0.08
5.51±0.30
0.10±0.00
5.28±0.36
Samples
pH
Volatile acidity (g/L acetic ac.)
°Brix
Sugars (%)
Vol. %
L*
a*
b*
Intensity (I)
Tone (N)
0.32±0.01
6.33±0.12
1.75±0.01
11.68±2.46
91.26±0.15
-1.62±0.01
11.70 ±0.06
0.27±0.01
3.13 ±0.10
Table 2. F values of all descriptors (24), samples (20), judges (9), replicate (R=3) and 2-way interactions Descriptors
Appearance: Colour Clarity Aroma: Citrus Apricot Alcohol Caramel Yeasty Floreal Fruity Stewed fruit Exotic fruit Woody Spicy Raisin Taste: Sour Sweet Salty Flavour: Apricot Fruity Woody Honey Raisin Mouthfeel: Astringent Burn
*** ** ** n.s.
F values Sample
Judge
66.82*** 16.09***
25.92*** 149.54***
72.11*** 11.24*** 4.71*** 48.66*** 9.58*** 5.73*** 5.96*** 14.95*** 4.35*** 4.34*** 1.58 n.s. 10.38***
SxJ
SxR
JxR
2.10 n.s. 23.29***
1.97*** 3.02***
2.26*** 2.73***
1.15 n.s. 1.41 n.s.
60.20*** 75.06*** 43.18*** 49.54*** 15.27*** 53.00*** 66.29*** 29.08*** 56.64*** 47.23*** 86.02*** 21.00***
0.18 n.s. 0.67 n.s. 0.81 n.s. 3.04* 0.08 n.s. 0.79 n.s. 0.77 n.s. 0.52 n.s. 0.21 n.s. 0.18 n.s. 0.94 n.s. 2.63 n.s.
3.67*** 3.03*** 2.32*** 2.73*** 2.05*** 2.68*** 2.11*** 2.67*** 2.99*** 2.23*** 2.68*** 1.94***
1.13 n.s. 0.89 n.s. 0.96 n.s. 1.51* 1.13 n.s 0.98 n.s. 0.74 n.s. 1.23 n.s. 1.12 n.s. 1.52* 0.84 n.s. 1.06 n.s.
1.21 n.s. 1.03 n.s. 1.66 n.s. 0.91 n.s 1.10 n.s. 1.14 n.s. 0.91 n.s. 1.53 n.s. 1.11 n.s. 1.48 n.s. 1.25 n.s. 1.24 n.s.
15.99*** 30.64*** 20.01***
24.03*** 31.60*** 27.64***
0.48 n.s. 0.61 n.s. 4.47*
2.66*** 2.32*** 3.62***
1.09 n.s. 1.13 n.s. 0.81 n.s.
1.20 n.s. 1.50 n.s. 1.05 n.s.
18.08*** 12.74*** 5.65*** 14.37*** 17.87***
92.14*** 64.17*** 35.85*** 64.45*** 25.36***
0.51 n.s. 0.94 n.s. 3.18 n.s. 0.25 n.s. 1.02 n.s.
3.24*** 2.53*** 2.29*** 2.43*** 2.67***
0.91 n.s. 1.00 n.s. 1.37 n.s. 0.78 n.s. 1.23 n.s.
2.01** 0.87 n.s. 1.21 n.s. 0.99 n.s. 1.32 n.s
12.71*** 9.64***
19.18*** 35.58***
3.14* 3.25*
2.59*** 3.14***
0.89 n.s. 1.29 n.s.
1.39 n.s. 1.68 n.s.
significant difference for significant difference for significant difference for no significant difference
p ≤ 0.001 p ≤ 0.01 p ≤ 0.05
Replicate
Figure 1. Sensory profile of the Mediterranean Malvasia wines of the 5 area.
F.woody
Colour Clarity
F.rasins A.alcohol 7 F.honey A.floreal F.apricot 4
A.fruity
F.fruity A.citrus
1
Salty
A.apricot
Sweet
A.exotic fruit
Acid
A.stewed fruit A.rasins
A.woody A.yeasty
Ar.caramel
2
L S I Cr Ca
Figure2. Bi-plot of Malvasia wine samples and physicochemical parameters. S2 F. woody
S6
a*
Astringent A. woody
S4 Colour intensity b*
Alcoholic content A. Alcohol Sour
S3
A. raisin A. citrus
L3 S5 A. yeasty L1
L2 S1
Ca2 Tritatable acidity
A. stewed fruit Burn L9 A. apricot F. spicy A. honey A. fruity L3 Cr1 L5 Cr3 Cr2 F. raisin L10 pH Volatile acidity A. floreal L6 F. fruity F.honey A. exotic fruit Sweet L1 L2 Brix L7 Sugar content L8 L4
Ca1 L*
3
Colour tone
Literature Cited 1. Agnolucci, M., Scarano, S., Santoro, S., Sassano, C., Toffanin, A., and Nuti, M. (2007): Genetic and phenotypic diversity of autochthonous Saccharomyces spp. strains associated to natural fermentation of ‘Malvasia delle Lipari’. Letters in Applied Microbiology 45, pp. 657–662. 2. Alves, R.F., Nascimento, A.M.D., Nogueira J.M.F. (2005): Characterization of the aroma profile of Madeira wine by sorptive extraction techniques. Analytica Chimica Acta 546, pp. 11–21. 3. A.O.A.C. (1990):Official Methods of Analysis of the A.O.A.C. (15th ed.), Kenneth Helrich (Ed.). Association of Official Analytical Chemists, Arlington, VA. 4. Bonino, M., Schellino, R., Rizzi, C., Aigotti, R., Delfini, C., Baiocchi, C.(2003): Aroma compounds of an Italian wine (Ruche) by HS .SPME analysis coupled with GC-ITMS. Food Chemistry, 80, pp. 125 .133. 5. Calò, A. (2005):“Le Malvasie”. Rivista di Viticoltura e di Enologia, 2-3-4, pp. 5-12. 6. Camara, J.S., Herbert , P., Marques , J.C., Alves M.A., (2004): Varietal flavour compounds of four grape varieties producing Madeira wines. Analytica Chimica Acta 513, pp. 203–207. 7. Costantini, V., Bellincontro, A., De Santis, D., Botondi, R., and Mencarelli, F. (2006): Metabolic changes of Malvasia grapes for wine production during postharvest drying. J. Agric. Food Chem., 54 (9), pp 3334-3340. 8. D.P.R. 20/9/1973: G.U. n. 28, 30/1/1974. 9. Farris, G.A., Zara, S., Zara, G. (2008): Invecchiamento biologico della Malvasia di Bosa (Sardegna). Rivista di viticoltura e di enologia, pp. 215-216. 10. Lanza, C. M., Scacco A., Mazzaglia A., Pagliarini E. (2008): Ricerca della identità sensoriale del vino Malvasia delle Lipari. Rivista di viticoltura e di enologia, 2-3-4, pp. 467-472. 11. Lo Iacono, V. (2008): Le attività vitivinicole in Sicilia e il ruolo delle isole minori. Rivista di viticoltura e di enologia, 2, 3, 4, pp. 299-309.
4
12. Lopez, R, Ortn, N, Prez-Trujillo, J.P., Cacho, J and Ferreira, V. (2003): Impact odorants of different young white wines from the Canary islands. J. Agric. Food Chem., 51, pp. 3419-3425. 13. Noble, A.C., R.A. Arnold, J. Buechsenstein, E.J. Leach, J.O. Schimdt, and P.M. Stern. (1987): Modification of a standardized system of wine aroma terminology. Am. J. Enol. Vitic. 38(2), pp. 143-146. 14. Ramos, R., Andrade, P. B., Seabra, R. M., Pereira, C., Ferriera, M. A., Faia, M. A. (1999): A preliminary study of non-coloured phenolics in wines of varietal white grapes (codega, gouveio and malvasia fina): effects of grape variety, grape maturation and technology of winemaking. Food Chemistry 67, pp. 39-44. 15.UNI 10957 (2003): Analisi sensoriale – Metodo per la definizione del profilo sensoriale degli alimenti e delle bevande. Ente Nazionale Italiano di Unificazione, Milano.
5
