Journal of Food Engineering 79 (2007) 758764 www.elsevier.

com/locate/jfoodeng

Assessment of colour and aroma in white wines vinications: Eects of grape maturity and soil type
Go mez-M guez, Manuela Go mez-M guez, M. Jose Isabel M. Vicario, Francisco J. Heredia *
Lab. Food Colour and Quality, Department of Nutrition and Food Science, Faculty of Pharmacy, University of Seville, 41012 Sevilla, Spain Received 19 October 2005; accepted 21 February 2006 Available online 5 May 2006

Abstract White wines were elaborated in an experimental winery from grapes (Vitis vinifera cv. Zalema) at dierent degrees of maturity and grown in dierent soil types. The eects of these factors on the quality characteristics, in terms of colour and aroma prole, of the wines at bottling were evaluated. Two types of soils (sand and clay) and two dierent harvest dates (early harvest and industrial harvest) were considered. With reference to colour, the highest colour dierences DE ab 2:37 1:05 were observed between wines from sandy soils and dierent harvest dates. Wines from sandy soil and early harvest date showed a more yellow-pale colour and lesser colour intensity. However, these dierences were hardly appreciated by tasters. On the contrary, eects of harvest date and soil type were more marked regarding the aromatic prole of the wines. Results showed that in order to elaborate a white wine of quality, in the case of grapes from clayey soils, the harvest date should not be postponed because it would cause a decrease of fresh aromas and an increase of ripe fruit notes, no favourable aromatic notes for a young white wine. 2006 Elsevier Ltd. All rights reserved.
Keywords: Aroma; Colour; Harvesting; Tristimulus colorimetry; Vineyard; White wine; Zalema

1. Introduction Colour is one of the most important sensory characteristics of wines, which inuences the consumers overall acceptability. Numerous factors can aect wine colour, such as the grape composition, the vinication method mez-Plaza, and the conditions of storage (Gil-Mun oz, Go nez, & Lo pez-Roca, 1997). It is, therefore, not surMart prising that the study of colour is important and it is a matter evolving constantly. On the other hand, aroma also inuences the organoleptic characteristics of the wines. Wine aroma is the result of the interaction between components of the grapes themselves and those produced during winemaking process, fermentation and aging. The wine aroma depends on many
*

Corresponding author. Tel.: +34 954556761; fax: + 34 954557017. E-mail address: [email protected] (F.J. Heredia).

factors: grape variety, environmental and management , practices, winemaking techniques, yeast, etc. (Falque ndez, & Dubourdieu, 2001). Ferna Grape composition, which will determine the sensory qualities of the wines, depends on intrinsic grape factors, such as the variety, as well as extrinsic factors, such as climatic conditions, soil, region and cultivation techniques (Dirninger et al., 1998; Jackson & Lombard, 1993; Jones & Davis, 2000; Troost, 1985). In this sense, some studies have tried to establish links between these viticultural parameters and wine composition. For example, previous studies have demonstrated the inuence of harvesting date rezon quality characteristics of wines. In this sense, Pe lez-San Jose (2006) found that the Magarin o and Gonza harvesting date (directly correlated with the degree of maturity of the grapes) inuences the colour characteristics of wines. Du-Plessis (1984) and Hamilton and Coombe (1992) have also reported that the moment for harvesting

0260-8774/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2006.02.038

 mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go

759

is an important factor in the making of quality wines. Regards to vineyard characteristics, several researchers have found that soil impacts the overall quality of the grape and thus the resultant sensory quality of the nished wines. Sayed (1992) and Wiebe and Anderson (1977), for instance found dierent wine compositions according to the soil type. Sabon, de Revel, Kotseridis, and Bertrand (2002) have also reported the inuence of Rhone Valley (France) soils on the aromatic composition of Grenache wines. And Wahl (1988) showed that Muller-Thurgau and Riesling wines, produced from vines growing on several soil types relocated to a common experimental site, could be discriminated from each other. This suggests that although macroclimate, mesoclimate, and fruit microclimate all have major impacts on wine quality and varietal typicity, soil may also have an independent eect. Vitis vinifera cv. Zalema is a white grape variety that grows exclusively in Huelva (southwest Spain), where it represents over 90% of the overall vineyard. It is used mainly to produce a typical regional white wine in the Spanish Denomination of Origin (DO) Condado de Huelva. There are few reference data concerning the com mez-M guez et al., in press; position of Zalema wines (Go n, & Recamales, 1999; RecaHernanz, Heredia, Beltra lez-Miret, & Hernanz, 2006), so conmales, Sayago, Gonza siderable research and development on the viticulture and oenology of this variety are still necessary. In this sense, the main objective of this paper is to study the inuence of some viticultural factors, such as harvest date and soil type, on the quality characteristics, colour and aroma, of Zalema white wines. This study is of practical interest because it provides some useful information for vine growers and winemakers in order to establish criteria to improve the quality of these wines. 2. Material and methods 2.1. Samples White wines were elaborated from grapes of the variety V. vinifera cv. Zalema, the main and more extensively grown variety in the Spanish Denomination of Origin (DO) Condado de Huelva, in the southwest of Spain. The grapes used in this study were grown, during the 2004 vintage, in four dierent vineyards in terms of soil

type (sand and clay). The chosen soils are the two majority soil types in the Condado de Huelva region. Two representative parcels for each soil type were selected. Two dierent harvesting dates were considered. One of them was the usual moment for harvesting (industrial harvest date) considered in Condado de Huelva region. This date is decided by the criteria of the winemaker according to chemical parameters, such as the weight of the berry and the density and titratable acidity of the musts. The other harvesting date was ten days before (early harvest date), which was chosen because winemakers from this wine-production area recently consider harvesting earlier to be important in order to obtain high quality wines. Between the two chosen dates, there was no signicant rainfall, and temperature and light levels were suitable for continued ripening, so the grapes remained turgid and no putrefaction was observed. Table 1 shows the chemical characteristics of the grapes at each harvest date and for the two studied soils. The wines were made in triplicate, so a total of 24 wines were submitted to study. Each wine was made following the traditional method of white winemaking from approximately 1000 kg of grapes. Thus, the grapes were manually harvested, placed in 20 kg boxes and transported to an experimental wine-production centre. They were destemmed, crushed and softly pressed in a horizontal press, where the juice was lightly sulted (0.02 g/L). The must obtained was transferred into a 600 L stainless steel tank, adjusting the total sulfur dioxide content of the must to 80 mg/L. In this tank, the must was kept at 12 3 C during 24 h. Then, residual solid parts were separated and the must was run straight into a fermentation tank, adjusting the total acidity to 6 g/L with tartaric acid. During fermentation processes, the temperature was kept at 18 2 C. No selected yeasts were used. At the end of fermentation (approximately after 3 weeks), wines were racked and they were stored in 300 L stainless steel tanks for the stabilization process at controlled temperature (12 3 C). Finally, the wines were ltered and bottled. Wine samples (250 mL) were taken at the moment of bottling and immediately analyzed. The samples were coded as follows: S1 (sandy soil, early harvest date), S2 (sandy soil, industrial harvest date), C1 (clayey soil, early harvest date), C2 (clayey soil, industrial harvest date).

Table 1 Chemical parameters for Zalema grapes at the two chosen harvest dates and for the two studied soils (sand and clay) Soil type Harvest date Weight of 100 berries (g) Sugar content (g/L)a Titratable acidity (g/L tartaric acid)a pHa
a

Sand 1 304 156 2.84 3.8 2 308 172 2.57 3.9

Clay 1 299 172 2.91 3.8 2 306 179 2.80 3.8

1 = early harvest date, 2 = industrial harvest date. a Values obtained as the mean of the two selected parcels for each type soil.

760

mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go

2.2. Analytical methods 2.2.1. Colour evaluation Colour measurements were made in a Hewlett Packard UVVis HP 8452 Spectrophotometer (Palo Alto, CA, USA), using 10 mm path length glass cells. Samples were ltered through Millipore-AP20 lters (Bedford, MA, USA) prior to the spectrophotometric analysis. The whole visible spectrum (380770 nm) was recorded (Dk = 2 nm) and illuminant D65 (daylight source) and 10 Observer (perception angle of a human observer) were used in the calculations as standard conditions. The CIELAB and CIELUV parameters L ; a ; b ; C ab ; hab ; suv were determined lvarez, Gonza by using the software CromaLab (Heredia, A rez, 2004), following the recommendations lez-Miret, & Ram of the Comission Internacionale of LEclairage (CIE, 1986). CIE, as well as several authors, consider that CIELAB parameters better dene the colour of wines and permit better dier ndez-Lo pez, & Lo pez-Roca, entiation (Almela, Javaloy, Ferna rez1996; Heredia, Troncoso, & Guzman-Chozas, 1997; Pe Magarin o & Gonzalez-San Jose, 2006). 2.2.2. Analysis of volatile compounds Quantitative analysis of volatile compounds was carried pez, Cacho, out by the procedure proposed by Ortega, Lo and Ferreira (2001) with slight modications. Nine milliliter of wine and 21 mL of water were salted with 10 g of (NH4)2SO4 and extracted with 0.6 mL of dichloromethane. One hundred microliter of a solution of 2-octanol (543 ppm) were used as internal standard. The obtained extract was analyzed by GC with FID detection. A HP6890 gas chromatograph equipped with an automatic sampler was used. The column (60 m 0.25 mm and 0.25 lm lm thickness) was a DB-Wax from J&W Scientic (Folsom, CA, USA). The temperature program was as follows: 40 C for 5 min, then raised at 2 C/min up to 170 C and nally raised at 1.5 C/min up to 200 C, by maintaining this temperature during ve minutes. Injector and detector were both kept at 250 C. Carrier gas was H2, the split ow was 39.3 mL/min, and the injection (3 lL) was performed in split mode (1:10). Quantitative data were obtained by interpolation of relative peak areas in the calibration graphs built by the analysis of synthetic wines containing known amounts of the analytes. 2.2.3. Sensory analysis The wine samples taken for analysis were also submitted to sensory evaluation by a panel of nine assessors experienced in wine tasting. The wines (70 mL, 10 C) were presented in coded standard glasses and covered with a watch-glass to minimize volatile components from escaping. Three wines were evaluated in each session. Colour and aroma attributes were judged. In this sense, panellists used a list of dierent colour and aroma attributes; for each sample, they had to choose the attributes perceived and the perceived intensity (in a scale of three points) of each attribute selected.

These sensory attributes had been previously selected by a trained panel. In this way, various sensory sessions, similar to that described above, were carried out by using a pre-established list of terms. Finally, on the basis of the tasters sensory experience, the list was reduced to the most important sensory attributes that describe Zalema wines. 2.3. Statistical analyses The statistical analyses were performed using software Statistica version 5.5 (StatSoft Inc., 1995). In order to study signicant dierences between the dierent types of wines in terms of colour and aroma characteristics, factorial analysis of variance was carried out using the general linear model procedure (GLM). Bonferronis test was used to evaluate the signicance of the analysis. 3. Results and discussion 3.1. Colour evaluation Fig. 1 shows the location of the wine samples analyzed in the (a*b*) colour plane in the CIELAB space as well as a representation of L* values versus C ab values (Fig. 1a and b, respectively). The CIELAB coordinates a* (red/ green), b* (yellow/blue) and L* (lightness) dene a samples colour. Lightness is the attribute of a visual sensation according to which a given visual stimulus appears to be more or less light, ranging from light to dark. C ab (chroma) and hab (hue), obtained from a* and b*, are psychophysical magnitudes. Hue (hab) is the attribute according to which colours have been traditionally dened as red, green, blue, yellow, etc. Chroma (C ab ) is the attribute of a visual sensation which permits a judgement to be made of the degree to which a chromatic stimulus diers from an achromatic stimulus of the same brightness (Wyszecki & Stiles, 1982). As can be seen in Fig. 1a, all the wine samples were situated in a narrow area near the origin of coordinates. a* values were found varying from 0.82 to 0.26 units, b* values were between 3.25 and 6.54 units, and hue values varied from 93 to 103. Besides, lightness, L* (Fig. 1b) was very high in all the samples (between 96.8 and 98.6 CIELAB units) and C ab values varied from 3.27 to 6.55 units. These colorimetric data indicate that samples presented yellow tones, light and very pale colour, with low chromatic intensity. These results are in agreement with those found in a previous paper for Zalema wines mez-M guez et al., in press) about the typical colour (Go of young white wines. It can be observed in Fig. 1 that S1 and S2 wines were found slightly separated from the other groups. However, in order to achieve a more accurate evaluation of the colour dierences between the four types of wines (S1, S2, C1 and C2), statistical evaluation of the data (CIELAB parameters and the saturation index in CIELUV space, s uv ) was carried out by applying the general linear model procedure. The model with xed eects involved soil type

 mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go

761

12
120

100 110

90

80

10
130

0 -8 -6 -4 -2 0 2 4

a*

100

(two levels), harvest date (two levels) and soil type harvest date interaction. The results obtained are shown in Table 2. The colour of the wines was signicantly aected by the harvest date (p 6 0.001) only for grapes grown in sandy soils. In this sense, harvesting later (industrial har vest) produced an increase of b ; C ab and suv , and a * decrease of L . Therefore, S1 wines, compared to S2 ones, had a bit lighter colour, less vivid and greener. In contrast, the harvest date did not show signicant inuence on wine colour for grapes from clayey soils. The soil type showed signicant eect on the colour of the wines at the industrial harvest date. In this sense, C2 wines, compared to S2 ones, showed a lesser colour inten sity (lower b ; C ab and suv values) and a greener colour (higher hab values). However, it is interesting to nd out if these signicant dierences can also be detected by the human eye. The parameter which provides this information is DE ab (colour dierences). This parameter is calculated as the Euclidean distance between two points in the three-dimensional space dened by L*, a* and b*: q 2 2 2 DE DL Da Db ab DE ab > 3 units indicates that the colour dierence between dierent samples can be easily detected by observers nez, Melgosa, Pe rez, Hita, & Negueruela, 2001). (Mart The highest value of DE ab was found between S1 and S2 wines DE 2 : 37 1 : 05 . This result indicates that a ab taster could hardly dierentiate the colour of these wines. Any other possible combination between groups of samples did not reach mean colour dierences higher than 3 units (the values obtained were between 0.18 and 1.45). In fact, in the sensory analysis carried out, the tasters assigned to all wines similar colour attributes (yellow-pale descriptors, slight green notes, and low colour intensity).

b* L*

99

98

97

96 3.2 4.0 4.8 5.6 6.4 7.2

3.2. Aromatic prole analysis The evaluation of the aromatic prole of the wines has been based on the study of the odour activity values (OAVs), because this aromatic index permits the valuation of the degree of participation of each compound in the nal aroma. In this sense, only compounds with OAVs > 1 are considered as active odorants. OAVs are calculated by

C*ab
b
S1 S2 C1 C2

Fig. 1. Representation of the wines in the colour diagram (a*b*) (a) and their L* values versus C ab values (b). S1: sandy soil, early harvest date; S2: sandy soil, industrial harvest date; C1: clayey soil, early harvest date; C2: clayey soil, industrial harvest date. Table 2 Colour of the wines at bottling (means SD) Soil type Harvest date L* a* b* C ab hab s uv Sand 1 98.37 0.03a 0.53 0.12a 3.70 0.34a 3.74 0.36a 98.06 1.09ab 0.06 0.01a 2 97.22 0.26b 0.48 0.08a 5.77 0.58b 5.79 0.57b 94.92 1.26a 0.09 0.01b Clay 1 97.75 0.17c 0.51 0.21a 4.42 0.36a 4.46 0.33a 96.80 3.19ab 0.07 0.01a

Eect 2 97.79 0.18c 0.68 0.13a 4.15 0.20a 4.21 0.17a 99.36 2.17b 0.07 0.01a Harvest date
***

Soil type
*

Interaction
***

ns
*** ***

ns
* * * *

ns
*** *** ** ***

ns
***

1 = early harvest date, 2 = industrial harvest date. On the same row, means with dierent letters dier signicantly. Signicance levels: ns = not signicant, *p 6 0.05, **p 6 0.01, ***p 6 0.001.

762

mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go

et al., 2001), where c using the equation OAV = c/t (Falque is the concentration of each compound in the wine and t is the olfactory perception threshold of the compound in wine. In this paper, t values have been taken from informa , Escudero, Cacho, & tion available in the literature (Cullere vant, 1991; Ferreira, Lo pez, & Cacho, Ferreira, 2004; Etie 2000; Guth, 1997; Peinado, Moreno, Bueno, Moreno, & Mauricio, 2004). Table 3 lists the OAVs of the volatile compounds identied and quantied in the dierent groups of wines (S1, S2, C1 and C2). Data correspond to the average of the six wines from each group. Each volatile compound has been associated to odour descriptors reported pez, on the literature (Ferreira et al., 2000; Guth, 1997; Lo ndez, & Cacho, 1999; Peinado et al., 2004). Ferreira, Herna On the basis of these odours descriptors, each compound has been also assigned to one or several aromatic series: solvent, oral, sweet, green (vegetal or herbaceous), fatty and fruity series, which were chosen on account of their extensive use for describing young table white wines (Peinado et al., 2004). As shown in Table 3, among 30 volatile compounds identied and quantitatively determined in the studied

wines, only 13 compounds were detected at levels above their odour thresholds (OAVs > 1). However, recent studies have reported the relevance in the overall aroma of substances at concentrations of at least 20% of its threshold unit (OAV > 0.2) (Belitz & Grosh, 1999; Rocha, Rodrigues, Coutinho, Delgadillo, & Coimbra, 2004). Thus, seven more volatile compounds (phenylethyl acetate, isobutanol, 1-hexanol, a-terpineol, methionol, linalool and c-butirolactone) were also considered as contributors to the aroma of these wines. Independently of soil type and harvest date, the active odorants found in the nal wines were, mainly, volatile fatty acids and their ethyl esters (Table 3). Esters contribute favourably to wine aroma as fruity notes, and fatty acids are associated to fatty notes (Rocha et al., 2004). Ethyl octanoate, followed by ethyl hexanoate, showed the highest OAVs compared to the rest of volatile compounds, these values being especially high in C2 wines (mean 213 and 71, respectively). Some higher alcohols, such as isoamyl alcohol, 2-phenylethanol and isobutanol, and two C6-alcohols (Z-3-hexenol and 1-hexanol) also had OAVs exceeding 0.2 units in all types of wines. Higher alcohols are reported

Table 3 Odour activity values (OAVs) of volatile compounds identied and quantied in Zalema white wines at bottling Aroma compound Type of winea S1 Ethyl octanoate Ethyl hexanoate Octanoic acid Hexanoic acid Isoamyl acetate Isovaleric acid Butyric acid Isoamyl alcohol Ethyl butyrate Decanoic acid Z-3-Hexenol 2-Phenylethanol Isobutyric acid Phenylethyl acetate Isobutanol 1-Hexanol a-Terpineol Methionol Linalool c-Butyrolactone Acetoin Ethyl lactate 1-Butanol Ethyl acetate Isobutyl acetate Furfural Ethyl 3-hydroxybutyrate Diethyl succinate Benzylalcohol Geraniol S2 C1 C2 Sweet, oral, fruity, banana, pear Fruity, apple, banana, brandy Fatty, rancid Cheese, fatty Banana, fruity, sweet Fatty, rancid Cheese Solvent Fruity Fatty, rancid Green Rose, honey Fruity, cheese, sweet Flowery Alcohol, nail polish Herbaceous, grass, woody, green Lilac, oral, sweet Baked cabbage Citrus, oral, sweet, grape-like Caramel, sweet, buttery Buttery, cream, owery Fruity, buttery Medicinal Pineapple, fruity, solvent Sweet, fruity, apple, banana Sweet Fruity Fruity, melon Herbaceous Rose, oral 2, 3, 6 6 5 5 3, 6 5 5 1 6 5 4 2, 3 5, 6, 3 2 1 4 2, 3 5 2, 3, 6 3, 5 5, 2 5, 6 1 1, 6 3, 6 3 6 6 4 2 Odour descriptors Odorant seriesb

119.32 10.01 150.89 15.67 145.39 23.44 212.98 16.16 51.63 1.82 55.75 1.94 55.77 6.07 71.48 9.78 17.27 2.42 17.40 1.40 20.04 4.82 18.55 0.26 13.37 0.89 14.87 1.20 16.77 3.78 11.21 1.38 14.09 0.002 11.03 2.44 14.34 2.28 13.19 0.63 10.39 0.09 10.70 0.21 10.52 0.20 12.69 1.90 5.89 0.36 6.26 0.33 5.84 0.44 7.74 1.94 3.83 0.30 5.16 0.56 5.24 0.80 4.05 0.42 2.29 0.25 2.36 0.18 2.34 0.14 2.67 0.12 1.49 0.22 1.32 0.23 1.71 0.34 2.08 0.30 1.41 0.04 1.48 0.12 1.51 0.09 1.08 0.05 1.01 0.07 1.71 0.06 1.42 0.63 1.24 0.16 1.00 0.05 1.04 0.01 0.99 0.07 0.96 0.06 0.39 0.11 0.69 0.16 0.70 0.37 0.60 0.23 0.36 0.02 0.42 0.02 0.40 0.09 0.36 0.03 0.27 0.03 0.29 0.003 0.32 0.02 0.24 0.004 0.18 0.07 0.20 0.04 0.26 0.07 0.18 0.02 0.18 0.03 0.24 0.07 0.23 0.13 0.16 0.02 0.19 0.01 0.20 0.007 0.21 0.01 0.19 0.004 0.16 0.01 0.18 0.02 0.19 0.03 0.17 0.02 0.02 0.004 0.02 0.003 0.02 0.006 0.02 0.01 0.04 0.01 0.04 0.009 0.08 0.044 0.05 0.03 0.003 0.0004 0.003 0.0007 0.003 0.0002 0.003 0.0008 0.008 0.0008 0.01 0.006 0.009 0.005 0.009 0.002 0.02 0.002 0.02 0.002 0.02 0.002 0.02 0.001 0.002 0.0003 0.0009 0.001 <LD 0.0009 0.001 0.004 0.0003 0.003 0.00005 0.003 0.00007 0.003 0.000 0.002 0.0002 0.003 0.0002 0.003 0.0004 0.003 0.0004 0.0003 0.000 0.0004 0.00001 0.0004 0.00002 0.0003 0.000 0.03 0.004 <LD 0.03 0.01 <LD

<LD: below the detection limit of the analytical method used. a S1: sandy soil, early harvest date; S2: sandy soil, industrial harvest date; C1: clayey soil, early harvest date; C2: clayey soil, industrial harvest date. b 1, Solvent; 2, oral; 3, sweet; 4, green; 5, fatty; 6, fruity.

 mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go

763

to contribute more to the intensity of the odour of the wine vant, 1991), while alcohols with six than to its quality (Etie carbon-atoms constitute a defect by depending on their , 2005). In this sense, only concentration (Calleja & Falque Z-3-hexenol exceeded slightly its odour threshold value (OAVs about 1.50). Finally, as can be seen in Table 3, other groups of volatile compounds, such as carbonyl compounds and terpenes, were found at levels well below their odour thresholds (OAVs < 1). Only some terpene compounds, such as a-terpineol and linalool, could be considered as odorant contributors because of their OAVs (higher than 0.2). c-butirolactone, that always occurs in wines (Rocha et al., 2004), was found in amounts of about 0.2 units in all wines, so it could also be considered as a compound with sensorial impact in Zalema wines. By comparing the four groups of wines (S1, S2, C1 and C2) on the basis of the OAVs (Table 3), most of volatile compounds showed the lowest OAV values for S1 wines (sandy soil, early harvest date), so they could be consider as the least aromatic wines. Nevertheless, in order to study possible signicant dierences between the wines in greater depth, OAVs listed in Table 3 for each individual compound were combined in their own aroma series. Thus, the global OAV for each series was obtained (Fig. 2). The data were subjected to a factorial analysis of variance by using the general linear model procedure (GLM) (Table 4). As can be seen in Fig. 2, all wines showed the same sequence in the aroma series; thus, the highest aroma contribution was that of the fruity series, followed by the sweet, oral, fatty, solvent and green series. The eect of harvest date, regarding the aroma of wines, was more marked with grapes cultivated in clayey soil (Table 4). In this sense, C1 and C2 wines had signicant dierences in all aroma series (solvent: p 6 0.05; oral, sweet, green and fruity: p 6 0.001), except in the fatty one. As can be seen in Fig. 2, the oral, sweet and fruity series were higher in wines elaborated with grapes harvested later, while the

Table 4 ANOVA of the aroma series in the wines studieda Series Harvest date eect S1S2 Solvent Floral Sweet Green Fatty Fruity
** *

Soil eect S1C1

C1C2
* *** *** ***

S2C2
* *** *** ***

ns ns ns ns

ns ns
*

ns
***

ns ns

ns
***

ns = No signicative interaction; *p 6 0.05, **p 6 0.01, ***p 6 0.001. a S1: sandy soil, early harvest date; S2: sandy soil, industrial harvest date; C1: clayey soil, early harvest date; C2: clayey soil, industrial harvest date.

solvent and green series exhibited lower OAVs. However, S1 and S2 wines only showed signicant dierences in the solvent (p 6 0.01) and oral series (p 6 0.05), these values being higher in S2 wines (Fig. 2). On the other hand, the soil type showed a signicant eect on the aroma of the wines, mainly in those of the industrial harvest date (Table 4). Regarding the early harvest date, wines from dierent soils (S1 and C1) only showed signicant dierences in the solvent (p 6 0.01) and green series (p 6 0.05), these OAVs being slightly higher in clayey soils (Fig. 2). However, wines from the industrial harvest date and dierent soils (S2 and C2 wines) showed signicant dierences (p 6 0.05 and p 6 0.001) in all aroma series, except the fatty one: C2 wines (wines from clayey soils) showed higher contributions of the oral, sweet and fruity series, while the solvent and green series decreased (Fig. 2). It is interesting to note that these wines were not considered as young white wines of high quality by the panel of tasters, which assigned aromatic descriptors of ripe fruit and freshness. These observations seem to indicate that wines with high OAVs in the fruity and sweet series, combined with lower OAVs in the green and solvent series, are not well evaluated as a typical young white wine of Zalema. 4. Conclusions It has been demonstrated that the harvest date (directly correlated with the degree of maturity of the grape) and the type of soil of the vineyard could aect both colour and aroma characteristics of the obtained Zalema wines. Although slight colour dierences were found statistically between the wines, these dierences were not appreciated by tasters. Regarding aromatic characteristics of wines, results have shown that in the case of clayey soils, within D.O. Condado de Huelva, the harvest date should not be postponed because of its negative impact on the aroma of the wines.

Fig. 2. Odour activity values (OAV) of odorant series for the wines studied (S1, S2, C1 and C2 wines). S1: sandy soil, early harvest date; S2: sandy soil, industrial harvest date; C1: clayey soil, early harvest date; C2: clayey soil, industrial harvest date.

References
ndez-Lo pez, J. A., & Lo pez-Roca, J. M. Almela, L., Javaloy, S., Ferna (1996). Varietal classication of young red wines in terms of chemical

764

mez-M guez et al. / Journal of Food Engineering 79 (2007) 758764 M.J. Go Bordeaux, France. American Journal of Enology and Viticulture, 51, 249261. pez, R., Ferreira, V., Herna ndez, P., & Cacho, J. (1999). Identication Lo of impact odorants of young red wines with Merlot, Cabernet Sauvignon and Grenache grape varieties: a comparative study. Journal of Agricultural and Food Chemistry, 79, 14611467. nez, J. A., Melgosa, M., Pe rez, M. M., Hita, E., & Negueruela, A. I. Mart (2001). Visual and instrumental color evaluation in red wines. Food Science and Technology International, 7, 439444. pez, R., Cacho, J., & Ferreira, V. (2001). Fast analysis of Ortega, C., Lo important wine volatile compounds. Development and validation of a new method based on gas chromatographic-ame ionisation detection analysis of dichloromethane microextracts. Journal of Chromatography A, 923, 205214. Peinado, R. A., Moreno, J., Bueno, J. E., Moreno, J. A., & Mauricio, J. C. (2004). Comparative study of aromatic compounds in two young white wines subjected to pre-fermentative cryomaceration. Food Chemistry, 84, 585590. rez-Magarin lez-San Jose , M. L. (2006). Polyphenols and Pe o, S., & Gonza colour variability of red wines made from grapes harvested at dierent ripeness grade. Food Chemistry, 96, 197208. lez-Miret, M. L., & Hernanz, D. Recamales, A. F., Sayago, A., Gonza (2006). The eect of time and storage conditions on the phenolic composition and colour of white wine. Food Research International, 39, 220229. Rocha, S. M., Rodrigues, F., Coutinho, P., Delgadillo, I., & Coimbra, M. A. (2004). Volatile composition of Baga red wine. Assessment of the identication of the would-be impact odorants. Analytica Chimica Acta, 513, 254262. Sabon, I., de Revel, G., Kotseridis, Y., & Bertrand, A. (2002). Determination of volatile compounds in Grenache wines in relation with dierent terroirs in the Rhone Valley. Journal of Agricultural and Food Chemistry, 50, 63416345. Sayed, H. (1992). Vineyard site suitability in Ontario. Ontario Grape and Wine adjustment Program, OMAFRA and Agriculture Canada. Publication N.10.92, Ministry of Agriculture and Food, Agriculture, Canada. StatSoft Inc. (1995). Statistica 5.5 for Windows Computer Program Manual, Tulsa, OK. StatSoft, Inc., 1995, 2300 East 14th Street, Tulsa, OK 74104. a del vino, S.A. Omega, Barcelona, Spain. Troost, G. (1985). Tecnolog Wahl, K. (1988). Climate and soil eects on grapevine and wine: Situation on the northern boarder of viticulture the example of Franconia. In Proceedings of second international symposium of cool climate viticulture and oenology (pp. 15). Auckland, New Zealand: New Zealand Soc. for Vitic. and Oenol. Wiebe, J., & Anderson, E. T. (1977). Site selection of grapes in the Niagara Peninsula. Vineland Station, Ontario: Horticultural Research Institute of Ontario. Wyszecki, G., & Stiles, W. S. (1982). Uniform color scales, in color science. concepts and methods. Quantitative data and formulae (2nd ed.). New York: John Wiley & Sons, Inc., p. 487.

and colour parameters. Journal of the Science of Food and Agriculture, 70, 173180. n: Belitz, H. D., & Grosh, W. (1999). Food chemistry (2nd ed.). Berl Springer-Verlag, pp. 319377. , E. (2005). Volatile composition of Menc a wines. Calleja, A., & Falque Food Chemistry, 90, 357363. , L., Escudero, A., Cacho, J., & Ferreira, V. (2004). Gas Cullere chromatographyolfactometry and chemical quantitative study of six premium quality Spanish aged red wines. Journal of Agricultural and Food Chemistry, 52, 16531660. Dirninger, N., Duc, D., Schneider, C., Dumas, V., Asselin, C., & Schaeer, A. (1998). Wine quality and terroirs: Inuence of environmental characteristics on the Gewurztraminer avor prole. Sciences des Aliments, 18, 193209. Du-Plessis, C. S. (1984). Optimum maturity and quality parameters in grapes: A review. South Africa Journal of Enology and Viticulture, 5, 3542. vant, P. X. (1991). Wine. In H. Maarse (Ed.), Volatile compounds in Etie foods and beverages (pp. 483546). New York, NY: Marcel Dekker. , E., Ferna ndez, E., & Dubourdieu, D. (2001). Dierentiation of Falque white wines by their aromatic index. Talanta, 54, 271281. pez, R., & Cacho, J. F. (2000). Quantitative determination Ferreira, V., Lo of the odorants of young red wines from dierent grape varieties. Journal of the Science of Food and Agriculture, 80, 16591667. mez-Plaza, E., Mart nez, A., & Lo pez-Roca, J. M. Gil-Mun oz, R., Go (1997). Evolution of the CIELAB and other spectrophotometric parameters during wine fermentation.Inuence of some pre and postfermentative factors. Food Research International, 30, 699705. mez-M guez, M. J., Gonza lez-Miret, M. L., Hernanz, D., Fe rna ndez, Go M. A., Vicario, I. M., & Heredia, F. J. (in press). Eects of prefermentative skin contact conditions on colour and phenolic content of white wines. Journal of Food Engineering, doi:10.1016/j.jfoodeng.2005. 09.021. Guth, H. (1997). Quantication and sensory studies of character impact odorants of dierent white wines varieties. Journal of Agriculture and Food Chemistry, 45, 30273032. Hamilton, R. P., & Coombe, B. G. (1992). Harvesting of wines grapes. In B. G. Coombe & P. R. Day (Eds.). Viticulture. Practices (Vol. 2, pp. 302327). South Australia: Underdale. lvarez, C., Gonza lez-Miret, M. L., & Ram rez, A. (2004). Heredia, F. J., A lisis de color. Registro General de la Propiedad CromaLab, ana Intelectual, SE-1052-04, Sevilla, Spain. n-Chozas, M. (1997). Heredia, F. J., Troncoso, A. M., & Guzma Multivariate characterization of aging status in red wines based on chromatic parameters. Food Chemistry, 60, 103108. n, R., & Recamales, M. A. F. (1999). Hernanz, D., Heredia, F. J., Beltra Optimization of an extraction method of aroma compounds in white wine using ultrasound. Talanta, 50, 413421. Jackson, D. I., & Lombard, P. B. (1993). Environmental and management practices aecting grape composition and wine quality A review. American Journal of Enology and Viticulture, 44, 409430. Jones, G. V., & Davis, R. E. (2000). Climate inuences on grapevine phenology, grape composition, and wine production and quality for