QUALITY EVALUATION OF WHITE AND RED WINE VARIETIES, FROM THE MAIN VINEYARDS OF ROMANIA

Abstract

The quality of any wine is intrinsically dependent on the quality and composition of the grapes used to produce it. In traditional winemaking countries such as Germany and France, wine quality is determined by geographic origin or the terroir of the wine. The aim of the present research is to determine the quality of wines from the main vineyards of Romania. In terms of quality rating, they display particular characters of the varieties, as well as the ecoclimatic conditions and ecopedological influence on the quality of wine. The work offers new information on the quality of the white wines obtained in main vineyards of Romania, useful for their promotion and marketing. The variation of the physico-chemical characteristics and elemental concentration represents a strong geological marker for wines geographical traceability.

Headnote

Received on 30th August 2017

Revised on 8th December 2017

The quality of any wine is intrinsically dependent on the quality and composition of the grapes used to produce it. In traditional winemaking countries such as Germany and France, wine quality is determined by geographic origin or the terroir of the wine. The aim of the present research is to determine the quality of wines from the main vineyards of Romania. In terms of quality rating, they display particular characters of the varieties, as well as the ecoclimatic conditions and ecopedological influence on the quality of wine. The work offers new information on the quality of the white wines obtained in main vineyards of Romania, useful for their promotion and marketing. The variation of the physico-chemical characteristics and elemental concentration represents a strong geological marker for wines geographical traceability.

Keywords: metals, quality, wines, Vitis vinifera

Introduction

According to the International Organization of Vine and Wine (OIV), wine is a food product exclusively obtained by total/partial fermentation from fresh grapes or the must obtained from pressed/impressed grapes. From the chemical point of view, wines are a complex beverage consisting of water, sugar, amino acids, ethanol, polyphenolic compounds, anthocyanins and organic/inorganic substances (Karatas et al., 2015).

The world of grapes and wines concerns at least 40 counties; the quality and the type of wines depend on natural ecoclimatic conditions and human factors. It said that, worldwide the climate of the different grape growing accounts for large parts of the diversity of varieties cultivated, quality and typenness of the wines and viticultural products (Bora et al., 2016).

Today the grapevine is cultivated worldwide. About 51% of the vine cultivated surfaces of the world is located in Europe, followed by Asia, America and Africa. In Romania the area planted with vines has been reduced since the 1990s, and currently it ranks 5 in Europe after Spain, France, Italy and Portugal. In 2013, Romania had 229 000 ha planted with vines (Bora et al., 2015a).

The wine quality is affected by both cultural and climatic factors, some of which are difficult to evaluate (Jackson et al., 1993). The quality of any wine is intrinsically dependent on the quality and composition of the grapes used to produce it. In traditional winemaking countries such as Germany and France, wine quality is determined by geographic origin or the terroir of the wine (Sequin, 1986). Terroir describes the relationship between an agricultural product and its geographical origin and considers that the region of production might influence the products characteristic. In the case of wine, terroir involves the interactions of grapevine, vineyard, ecoclimatic conditions and human factors such as viticultural and oenological practices. More specifically, mesoclimatic variability has to be taken into account, as well as altitude, inclination, orientation and composition of soil. Soil is one of the most important factors of the production area which shows a particular interest for the assessment of the environmental effects on the mineral composition of the wine. In the ongoing effort to develop new monitoring techniques of the wine, geochemical marks significantly improve the traceability of wines to their origins, especially the mineral compositions of wines. Together, the various aspects of terroir affect the development and composition of the grapes, which in turn influences the characters of the wine, so terroir can be seen as a proxy for wine quality.

The aim of our research was therefore to (1) evaluate the quality of white and red wine varieties from the main vineyards of Romania, (2) determination of total metal concentration from wines samples (3) establishing some Pearson correlation coefficient between quality parameters of wine and between the determined elements, (4) determination of wine geographical traceability amongst physicochemical parameters and metal concentration from wines samples.

Materials and methods

Study area

A total of 84 wine samples were analyzed (2 red wines varieties (Fetească neagră, Merlot) and 2 white wines varieties (Sauvignon blanc, Fetească regală)). Samples originated from five different Romanian vineyard: Dealu Bujorului vineyard (45°52'10" N-27°55'8"E) (n = 24), Murfatlar vineyard (44°10'25" N-28°24'30" E) (n = 12), Tarnava vineyard (46°10'31" N-23°54'52" E) (n = 12), Iasi vineyard (47°9'44" N-27°35'20" E) (n = 12) and Ştefăneşti-Argeş vineyard (44°51'53" N24°57'0" E) (n = 24). The regions differ in geographical features and also by soils geological/pedological patterns. The Dealu Bujorului region is characterized by an alternate landscape, from flat to hilly areas, with altitude between 100 and 225 m and the predominant soil is levigated chernozem with a clayey sand texture and pH of 7.5 - 8.0. The Murfatlar region landscape is characterized by high fragmentation; with altitude between 0 and 100 m and the predominant soil is chernozem having clayey sand texture with pH values between 7.6 and 8.3. The Tarnava vineyard is situated on the southern slopes and stating at the altitude of 250-270 m up to 400450 m, the slope of these lands being between 15-35%. The largest slopes are situated on the river Tarnava Mare and descend on Tarnava Mica, Mures and the inner valleys. Predominant soil was aluvisol with pH between values 7.6 and 8.3. Iasi vineyard are represented by fragmented hills with low sloping plateaus, as a result of the erosion action with altitude between 100 and 200 m and the predominant soil is chernozems with pH values between 5.3-8.0. On the other hand, the landscape of Ştefaneşti-Argeş vineyard is very billowy terrain with altitude between 200 and 412 m and the predominant soil is protisols with pH 6.67.9.

Sample collection and microvinification process

The wine samples used were obtained from Sauvignon blanc, Feteasca regala, Feteasca neagra and Merlot wines under the conditions of 2015 and 2016 year, from Dealu Bujorului, Murfatlar, Tarnava, Iaşi and Ştefaneşti-Argeş vineyard. Around 4-5 kg of grapes / cultivar was collected from 10 vines / replication. Three repetitions / cultivar were used, placed in randomized blocks. The wine samples resulted from micro-wine production. Micro-vine production it was done according to the methodology described by Bora et al. (2016). All wines were providing by the wineries as finished wines in 750 mL glass bottles with cork stoppers and were stored at 3-4 0C before analysis. One bottle was used for each sample, and three replicates were taken. All the vines were planted from 1979, and the vine plantation was organized with 2.2 x 1 m distance between plants and rows. Vines were planted according to the Guyot system and were grown on speliers.

Reagents and solutions

All chemicals used in the experiment were of analytical grade. For the physicochemical analyses of wine we used Sigma-Aldrich chemicals. High purity ICP-MS multielement standard solution XXI CertiPUR obtained from Merck was used for calibration curve in the quantitative analysis. Standard solution of metals K, Na, Ca, Mg, Fe, Cu, Mn, Zn and Li 1000±0.002 mg/L were used. The control sample and working standards were freshly prepared for the experiments out of the stock solution. The accuracy of the methods was determined by running replicate analyses and the obtained values ranged, for different elements, between 0.8 and 13.1%. The global recovery of the elements varied between 85.6-101.3%.

Blanks and triplicate samples (n = 7) were considered in the experiment for each procedure. The variation coefficient was under 4%. The calibration curves were used to determine the detection limits (ppb) for each method. Limit of quantification (LoQ) and Limit of detection (LoD) were calculated as follows: LoQ = 10 SD/s and LoD = 3SD/s, where SD is the estimation of the standard deviation of the regression line while s is the slope of the calibration curve.

Sample preparation for determination of heavy metals from wine

The physico-chemical analyses of wine were performed in the Laboratory of Winemaking of the RSDVV Bujoru; these methods of analysis were described in (Postolache et al., 2016). In order to determine the heavy metals from wine samples, a volume of 0.2 mL wine was mixed with 1 mL H2O2 and 7 mL HNO3 69% and after 15-30 minutes the mineralization was performed in three steps using a microwave system Milestone START D Microwave Digestion System as follows: step I - 10 min. at 200°C, step II - 15 min. at 200°C and step III - 60 min. under ventilation at 35°C. Afterwards all samples were filtered through a 0.45 mm filter and diluted to 50 mL.

Equipments

In this case the pH was measured with WTW inoLab pH 7110. For determination of physico-chemical characteristics of red wine we used UV-VIS spectrometer (SpectronicHelios Gamma UV-Vis, ThermoFisher Scientific). A mass spectrometer with inductively coupled plasma (ICP-MS) iCAP Q Thermo scientific was used to determine the metals. The experimental conditions used for measurement were: argon flow on nebulizer (0.84 L/min.), auxiliary gas flow 0.80 L/min., argon flow in plasma 15 L/min., lens voltage 7.31 V; RF power in plasma 1100 W, spray chamber temperature (2.51±1.00 °C). Accuracy was calculated for the elements taken into consideration (0.5-5.0%).

Statistical analysis

The statistical interpretation of the results was performed using the SPSS, version 24 (SPPS Inc. Chicago, IL, USA). The following statistical parameters were determined: average error, standard deviation, arithmetic average, using the SPSS version 24 (SPPS Inc. Chicago, IL, USA). In order to determine whether the main quality parameters of wine can influence each other, the correlation coefficient was calculated using SPSS version 23 Pearson (SPSS Inc., Chicago, IL., USA). Linear discriminant analysis (LDA) was performed in order to separate the wines by region and to indentify the markers with a significant discrimination value (variables with Wilk's lambda near zero, p values <0.005 and higher F coefficients). Linear discriminant analysis (LDA) was performed using Microsoft Excel 2016 and XLSTAT Addinsoft version 15.5.03.3707.

Results and discussion

Physico-chemical characteristics of wines samples

When analyzing each wine variety can be noted that the obtained wine presented are variable alcohol content. Vine varieties for red wine recorded the highest values for alcohol content [15.51±0.38 (% vol.); 15.40±0.14 (% vol.) Merlot Dealu Bujorului and 14.89±0.38 (% vol.); 14.71±0.10 (% vol.) Feteasca neagra Ştefaneşti-Argeş vineyard], vine varieties for white wine recorded the highest values for alcohol content in Sauvignon blanc growing in Ştefaneşti-Argeş vineyard [13.37±0.17 (% vol.) 2015; 12.54±0.12 (% vol.) 2016], followed by Sauvignon blanc from Dealu Bujorului vineyard [12.41±0.19 (% vol.) 2015; 12.48±0.03 (% vol.) 2016]. The lowest values of alcoholic strength were recorded in varieties grown in Murfatlar vineyard. The analyzed wines have a high alcoholic potential, being within the normal range between (10.71 to 13.37 % vol. white wines) and (13.49 to 15.51 % vol. red wines). The results are comparable with the results reported by de Bruijn et al., 2014 (13.30±0.10 [% vol.], 12.30±0.10 [% vol.] Sauvignon Blanc wines), Miličević et al., 2014 (13.20±0.10 [% vol.] Syrah), Trig°Córdoba et al., 2015 (13.60±0.00 [% vol.] Godello [white wine], and Baiano et al., 2015 (11.44±0.11 [% vol.] Nero din Troia [red wine]).

Regarding total acidity (g/L C4H8O6), vine varieties for white wines grown in Târnava vineyard Sauvignon blanc [7.50±0.01 g/L CäO6 (2015)] and Iaşi vineyard Feteasca regala [7.70±0.22 g/L CäO6 (2016)] recorded the highest values for total acidity, while varieties grown in Dealu Bujorului, Murfatlar and Ştefaneşti-Argeş recorded lower values. On the opposite pole Feteasca neagră from Dealu Bujorului [9.19±0.09 g/L CÄO6 (2015)] and Merlot [8.83±0.01 g/L CÄO6 (2015)] recorded the highest values, followed by Merlot from Ştefaneşti-Argeş [7.28±0.07 g/L C4H8O6 (2016)]. The results are comparable with data obtained by Budić-Leto et al., 2009 (4.68±0.03 [g/L C4H6O6]), and Miličević et al., 2014 (5.50 ± 0.15 [g/L C4H6O6] Syrah).

Varieties grown in Murfatlar vineyard recorded the lowest values for volatile acidity [0.31±0.02 g/L CH3COOH (2015); 0.38±0.03 g/L CH3COOH (2016) recorded by Sauvignon blanc and 0.33±0.02 g/L CH3COOH (2015); 0.36±0.03 g/L CH3COOH (2016) recorded by Feteasca regala], while Sauvignon blanc from Ştefaneşti-Argeş [0.51±0.01 g/L CH3COOH (2015)] recorded de highest values, from white wines. Merlot varieties from Dealu Bujorului [0.72±0.01 g/L CH3COOH (2015); 0.65±0.03 g/L CH3COOH (2016)] and Feteasca neagra from Ştefaneşti-Argeş [0.62±0.01 g/L CH3COOH (2015); 0.63±0.02 g/L CH3COOH (2016)] recorded the highest values. Between the analyzed variants there are significant differences. The results are comparable with those reported by Bruijn et al., 2014 (0.50±0.00 [g/L CH3COOH], 0.40±0.00 [g/L CH3COOH] Sauvignon Blanc wines), Miličević et al., 2014 (0.46±0.20 [g/L CH3COO]) Syrah), and greater than those obtained by Baiano et al., 2015 (0.12±0.01 [g/L CH3COO]).

In case of non-reducible extract, the lowest values was recorded by Sauvignon blanc from Ştefaneşti-Argeş [19.07±0.25 g/L (2015); 19.09±0.30 g/L (2016)], followed by Sauvignon blanc from Târnavelor vineyard [19.48±0.16 g/L (2015)] and Feteasca regala from Iaşi [19.65±0.06 g/L (2016)]. Regarding red wines, Merlot variety from Ştefaneşti-Argeş [24.54±0.41 g/L (2015); 25.58±0.34 g/L (2016)] recorded the lowest values. On the opposite pole Sauvignon blanc (2015), Feteasca regala (2015) and Feteasca neagra (2015), Merlot (2015) from Dealu Bujorului recorded the highest non-reducible extract. The results are comparable with those reported by Bora et al., 2016 (29.00±0.50 [g/L Muscat Ottonel], 20.00±1.00 [g/L Feteasca alba], 27.0±01.00 [g/L Sauvignon blanc], 21.10±1.00 [g/L Aligoté]).

The sugar content values vary within large range [Dealu Bujorului, Sauvignon blanc 10.49±0.15 g/L (2015); Feteasca neagra 10.40±0.10 g/L (2016)] to [Dealu Bujorului, Merlot 36.89±0.05 g/L (2015); 10.40±0.04 g/L (2016)]. The lowest values were recorded by Feteasca regala from Ştefaneşti-Argeş [1.48±0.02 g/L (2015); 1.17±0.15 g/L (2016)] to [0.73±0.64 g/L (2016) Feteasca regala from Iaşi vineyard]. The wines under study fall into the category of dry, semi-dry and semisweet wines. The results are comparable with those reported by Bora et al., 2016 (30.70±0.75 [g/L Muscat Ottonel], 23.00±1.00 [g/L Şarba], 12.0±5.00 [g/L Sauvignon blanc], 72.00±1.00 [g/L Italian Riesling]), and higher than Miličević et al., 2014 (2.85±0.25 [mg/L, Syrah].

The highest pH was obtained in the wine produces from Sauvignon blanc from Murfatlar vineyard [3.66±0.01 (2015)], followed by Sauvignon bl2anc from Ştefaneşti-Argeş vineyard [3.75±0.01 (2015)] and Merlot blanc from ŞtefaneştiArgeş [3.77±0.02 (2015)]. Feteasca regala variety from Iaşi vineyard recorded the lowest pH values [2.86±0.14 (2016)] and Feteasca neagra variety from ŞtefaneştiArgeş vineyard [3.07±0.01 (2015)]. The results are comparable with those reported by Bora et al., 2016 (3.47±0.10 [Muscat Ottonel], 3.32±0.17 [Şarba], 3.54±0.17 [Sauvignon blanc], 3.46±0.01 [Italian Riesling]), and also with those reported by Bora et al., 2016b (3.30±0.01 [Merlot].

In the case of free SO2, white wines recorded the highest values at Feteasca regala [40.71±0.02 mg/L (2015) Murfatlar vineyard], followed by the Sauvignon blanc [39.50±1.57 mg/L (2016) Murfatlar vineyard] and at the opposite pole with the lowest free SO2 was Feteasca regala [21.40±0.84 mg/L (2015) Ştefaneşti-Argeş vineyard] followed by the Feteasca regala [22.61±1.51 mg/L (2015) Iaşi vineyard]. By comparing vines for white wine with varieties of red wine, one can get tired of the fact that the vines for red wines have a low content of free SO2. The wines obtained in Ştefaneşti-Argeş vineyard recorded the highest [Feteasca neagra 31.26± 1.28 mg/L (2015)] followed by the [Merlot 31.62±0.28 mg/L (2015)]. Analyzing the results in terms of free SO2 content, it can be seen that all wines have a much lower content than the one required by low, therefore the wine can be preserved or consumed.

The highest total SO2 content was recorded in wines from Murfatlar vineyard [Sauvignon blanc 179.13±0.61 mg/L (2015); Sauvignon blanc 138.84±9.41 mg/L (2016)] and [Feteasca regala 179.45±1.73 mg/L (2015)] followed by the Sauvignon blanc [137.00±4.26 mg/L (2016) Ştefaneşti-Argeş vineyard]. The Sauvignon blanc variety from Ştefaneşti-Argeş vineyard [50.44±1.86 mg/L (2015)] and Feteasca regala from Iaşi vineyard [64.83±1.36 mg/L (2015)] were recorded the lowest concentration of total SO2. In this case, the red wine also recorded lower values. The highest concentration of total SO2 was recorded in Merlot [85.11±1.51 mg/L (2016), Dealu Bujorului vineyard and 81.13±2.40 mg/L (2016) Ştefaneşti-Argeş vineyard]. The results are comparable with those reported by Bora et al., 2016 (60.00±1.20 [mg/L Free SO2 Muscat Ottonel], 52.32±0.60 [Şarba mg/L Free SO2], 20.00±0.61 [Sauvignon blanc mg/L Free SO2], 18.00±0.30 [Italian Riesling mg/L Free SO2]), and by Bora et al., 2016 (240.00±6.40 [mg/L Total SO2 Muscat Ottonel], 217.00±1.00 [Şarba mg/L Total SO2], 163.00±1.00 [Sauvignon blanc mg/L Total SO2], 208.00±2.00 [Italian Riesling mg/L Total SO2]).

Regarding the color intensity of the tested wines, based on the results, we can state that the highest color tint was recorded in the Feteasca neagra variety from Dealu Bujorului vineyard [9.097±0.057 (2015); 8.941±0.042 (2016)], compared to the same variety grown in Ştefaneşti-Argeş vineyard [8.320±0.026 (2015); 7.865±0.069 (2016)]. The results are comparable with those reported by Bora et al., 2016b (8.740±0.060 [Merlot], 7.700±0.030 [Cabernet Sauvignon], 8.380±0.090 [Feteasca Neagra]).

The highest concentration color tint was recorded in Feteasca neagra from Ştefaneşti-Argeş vineyard [0.877±0.350 (2015) and 0.817±0.021 (2016)] compared to the same variety grown in Dealu Bujorului [0.793±0.001 (2015) and 0.784±0.002 (2016)]. Regarding Feteasca neagra, Merlot variety from Dealu Bujorului vineyard and Merlot variety from Ştefaneşti-Argeş vineyard there is not statistical difference between the analyzed variants. The results are comparable with those reported by Bora et al., 2016b (0.810±0.040 [Merlot], 0.690±0.020 [Cabernet Sauvignon], 0.740±0.020 [Feteasca Neagra]).

In case of total polyphenols, the lowest values were recorded by wine from Ştefaneşti-Argeş vineyard [Feteasca neagra 0.88±0.04 g/L (2015) and 2.08±0.17 g/L (2016)]; [Merlot 0.78±0.01 g/L (2015) and 1.94±0.06 g/L (2016)] compared to the same variety grown in Dealu Bujorului vineyard [Feteasca neagra 2.04±0.02 g/L (2015) and 2.26±0.04 g/L (2016)]; [Merlot 2.08±0.02 g/L (2015) and 2.02±0.07 g/L (2016)]. The results are comparable with those reported by Bora et al., 2016b (1.29±0.01 [g/L Merlot], 1.14±0.02 [g/L Cabernet Sauvignon], 1.28±0.01 [g/L Feteasca Neagra]).

Just like in case of total polyphenols, the anthocyanis recorded the lowest values at wine from Ştefaneşti-Argeş vineyard [Feteasca neagra 407.92±1.25 g/L (2015) and 489.62±19.71 g/L (2016)];'[Merlot 316.81±0.61 g/L (2015) and 418.33±16.96 g/L (2016)] compared to the same variety grown in Dealu Bujorului vineyard [Feteasca neagra 913.67±4.04 g/L (2015) and 621.87±11.99 g/L (2016)]; [Merlot 663.44±7.05 g/L (2015) and 502.56±6.88 g/L (2016)]. The results are comparable with those reported by Bora et al., 2016b (302.67±2.08 [g/L Merlot], 216.33±1.53 [g/L Cabernet Sauvignon], 281.33±1.53 [g/L Feteasca Neagra]).

Total metal concentration from wines samples

As expected potassium was the most abundant element in all investigated red and white wine samples since this element is essential for the growth and development of plants and is often a component of fertiliser (Rodrigues et al., 2011). The vine requires high contents of potassium for its mineral nutrition, which can be further found in must and wine. According to Ţârdea et al. (2001) Potassium is responsible for the finesse of the wine, while the low potassium samples have a harsh taste. The highest concentration of Potassium was recorded in the wine samples from Dealu Bujorului vineyard [Merlot 485.79±2.14 mg/L (2015); 489.38±0.21 mg/L (2016)] followed by Feteasca neagra [335.97±7.09 mg/L (2015); 326.70±4.99 mg/L (2016)] and Merlot [291.12±5.49 mg/L (2015)] from Ştefaneşti-Argeş vineyard. Among the analyzed variants was a very significant difference (F = 54.115; p < 0.000). The polyfactorial analysis indicated that the area of vineyard culture significantly influences the accumulation of K in wines (F = 5.732; p < 0.000), while the variety and the interaction between area of culture and had no significant influence on this character. These results are lower compared to the values reported in the literature (Iglesias et al., 2007 - average values of 819.61 mg/L; Álvarez et al., 2012 - average values of 865.30 mg/L), and agree with those reported by Bora et al., 2008 [Feteasca alba 323.26±3.25 mg/L (2014)], [Feteasca regala 235.86±10.25 mg/L (2014)].

The highest Sodium concentration was found in wine sample from Dealu Bujorului vineyard [Feteasca neagra 51.82±0.98 mg/L (2016)] followed by wine from Murfatlar vineyard [Feteasca regala 46.01±1.32 (2015)].

Feteasca regala variety from Dealu Bujorului vineyard [22.41±0.90 mg/L (2015)] and Merlot variety from Ştefaneşti-Argeş vineyard [25.55±1.49 mg/L (2015)]. When comparing the average value (37.37 mg/L Na) to the ones reported in the legislation, one can notice that the concentrations of Na are below the allowed maximum limit (60 mg/L). The Na content in our study are similar with the results published on Serbian (Ražić and Onjia, 2010 average values of 29.65 mg/L Na), Czech (Kment et al., 2005 average values of 14.7 mg/L Na) and Spanish (Iglesias et al., 2007 average values of 37.19 mg/L Na) wines.

The large amounts of calcium present in wines can be due to some exogenous sources, treatment with bentonite, filtration with alluvial infusorial soil (diatomite), storage of wine in concrete tanks, and de-acidification of calcium carbonate. Low temperature and pH values of 2.9-3.2 favor TCa crystals formation because tartrate anions (T2-) rise when combines with the calcium in wine. On the other hand, at high temperature the formation of calcium malate is favored (Bora et al., 2015b).

The highest concentration of Ca was recorded in wine from Iaşi vineyard by Sauvignon blanc variety [78.22±3.64 mg/L (2015); 78.47±1.27 mg/L (2016)] and Feteasca regala variety from Ştefaneşti-Argeş vineyard [82.34±1.92 mg/L (2015)]. In case of Mg concentration, wine from Târnavelor vineyard [Sauvignon blanc 144.42±2.71 mg/L (2016)], from Iaşi vineyard [Feteasca regala (131.15±6.98 mg/L (2015)] and Ştefaneşti-Argeşi vineyard [Sauvignon blanc 145.29±3.82 mg/L (2016) and Feteasca regala Ш.49±3.74 mg/L (2015); 141.95±8.02 mg/L (2016)] recorded the highest concentration of Mg.

The values obtained for the Mg and Ca contents in our selected wines were in good agreement with the results for Macedonian (Ivanova-Petropulos et al., 2013 average values of 83.5 mg/L Ca and 98.20 mg/L Mg), Serbian (Ražić and Onjia, 2010 - average values of 37 mg/L Ca and 95.73 mg/L Mg), Croatian (Vrček et al., 2011 - average values of 65.90 mg/L Ca and 68.70 mg/L Mg) and also Czech wines (Kment et al., 2005 - average values of 108.00 mg/L Ca and 75.40 mg/L Mg). On the other hand, our Ca and Mg contents were significantly higher than published data for wines from Argentina (Lara et al., 2005 average values of 12.50 mg/L Ca) and Belgium (Coetzee et al., 2014 average values of 6.73 mg/L Ca and 12.05 mg/L Mg).

Sauvignon blanc from Dealu Bujorului vineyard [3.49±0.54 mg/L (2015); 3.58±0.43 mg/L (2016)], from Murfatlar vineyard [3.58±0.43 mg/L (2016)] and also from Ştefaneşti-Argeş vineyard [3.25±0.67 mg/L (2016)] recorded the highest concentration of Fe. In terms of Fe concentration in red wine, the recorded concentration are within normal limits, Dealu Bujorului vineyard [Feteasca neagra 1.86±0.62 mg/L (2015); 1.74±0.10 mg/L (2016); Merlot 2.10±0.65 mg/L (2015); 2.13±0.01 mg/L (2016)] and Ştefaneşti-Argeş [Feteasca neagra 1.86±0.62 mg/L (2015); 2.92±1.01 mg/L (2016); Merlot 2.11±0.32 mg/L (2015); 1.90±0.69 mg/L (2016)].

Cu concentration was within wide limits, recorded the highest concentration of Cu were recorded in wine from Dealu Bujorului vineyard [Feteasca regala 0.92±0.03 mg/L (2016)] and in wine from Murfatlar vineyard [Feteasca regala 0.94±0.03 mg/L (2016)]. Sauvignon blanc variety from Târnava vineyard [3.36±0.05 mg/L (2015); 0.35±0.05 mg/L (2016)] and Feteasca regala from Iaşi vineyard [0.31±0.08 mg/L (2015)] recorded the lowest concentration of Cu in wine sample. Anyway, the values of Na concentration are below the maximum limit allowed by the applicable law (1 mg/L).

For the nutrition of vines, manganese is a microelement that it takes from soil and accumulates in grapes at very low concentrations. The highest concentration of Mn in wine were recorded in wine from Dealu Bujorului vineyard [Feteasca regala 0.61±0.09 mg/L (2016); Feteasca neagra 0.52±0.15 mg/L (2016)], followed by Feteasca regala from Murfatlar vineyard [0.61±0.09 mg/L (2016)], Feteasca regala from Tarnava vineyard [0.54±0.13 mg/L (2016)] and also Feteasca regala from Ştefaneşti-Argeş vineyard [0.58±0.08 mg/L (2016)]. Low concentration were recorded in wine from Dealu Bujorului vineyard [Feteasca neagra 0.17±0.02 mg/L (2015); Merlot 0.21±0.08 mg/L (2015) and 0.25±0.06 mg/L (2016)] and Murfatlar vineyard [Sauvignon blanc 0.29±0.06 mg/L (2015) and 0.27±0.02 mg/L (2016)]. Statistical analysis indicated significant differences between the analyzed variants (F = 2.828;p < 0.041).

The highest concentration of Zn in wine were recorded in wine from Dealu Bujorului vineyard [Feteasca regala 3.16±0.05 mg/L (2015); 3.06±0.09 mg/L (2016)] and wine from Murfatlar vineyard [Feteasca regala 4.01±0.19 mg/L (2015)]. Sauvignon blanc variety from Ştefaneşti-Argeş [1.05±0.16 mg/L (2015); 1.00±0.02 mg/L (2016)], Feteasca regala [1.11±0.09 mg/L (2015)] and Merlot [0.73±0.16 mg/L (2015)] from the same vineyard recorded the lowest concentration of Zn in wine sample. Zn concentration was within wide limits [4.01±0.19 mg/L maximum value] to [0.73±0.16 mg/L minimum value] with an average value of [1.08 mg/L Zn]. Based on the statistical analysis, it can be observed that the between analyzed variants are distinctly significant difference (F = 17.550; p = 0.000). The average value (1.85 mg/L) of Zn concentrations is below the maximum limit allowed by the law (5 mg/L).

The behavior of Li in wine resamble that of alkaline-earth metals and particularly the one of Mg. When aging the bottled wine, a reducing environment is created, causinf the lithium to be expelled out of the wine. The highest concentration of lithium in wine were recorded in wine out of Dealu Bujorului vineyard [Feteasca regala 14.15±0.47 mg/L (2015); Feteasca neagra 14.67±0.35 mg/L (2015); 14.98±1.17 mg/L (2016)], wine from Murfatlar vineyard [Feteasca regala 13.58±1.04 mg/L (2015)], wine from Iaşi vineyard [Feteasca regala 13.09±1.35 mg/L (2016)] and also wine from Ştefaneşti-Argeş vineyard [Sauvignon blanc 13.31±1.02 mg/L (2015); 13.3Ш.68' mg/L (2016)], '[Feteasca neagra 14.00±0.23 mg/L (2015); 13.03±2.53 mg/L (2016)]. Low concentration were recorded in wine from Iaşi vineyard [10.71±1.44 mg/L (2015)] followed by wine from Târnava vineyard [Sauvignon blanc 11.31±0.98 mg/L (2015); 11.39±1.96 mg/L (2016)].

Cu, Mn, Zn, and Li were also present in amounts similar to previously published results (Pohl 2007; Fabani et al., 2010; Di Paola-Naranjo et al., 2011; IvanovaPetropulos et al., 2013; Avram et al., 2014; Catarino et al., 2014; Geana et al., 2016).

This higher content of some metals may be due to the viticultural practices, the use of fertilizers for cultivation of vine (K, Ca, Cu) the winemaking process or addition of substances for wine clearing as bentonite (Na, Ca, Fe). Cu content is below the limit of detection due to the modern technology for obtaining wines in a controlled manner.

The Pearson correlation between the main parameters analysed in wine In order to determine whether the main quality parameters of wine can influence each other, the Pearson correlation coefficient was calculated for each studied parameter as it shown in Table 4 and 5. A Pearson correlation coefficient value higher than 0.5 shows a strong correlation between the analysed varieties, a positive correlation between the two parameters shows that both parameters increased, a negative correlation indicates that a parameter increased while the second one decreased and vice-versa.

These provide a large number of both positive and negative correlations between the main parameters of the analysed wines. There are some relevant examples: Alcohol & Total acidity, (r2 = 0.615 ··); Alcohol & Volatile acidity (r2 = 0.757··); Alcohol & Non-Reducible extract (r2 = 0.869··); Alcohol & Coloring intensity (r2 = 0.876··); Alcohol & Color tint (r2 = 0.874··); Alcohol & Total polyphenols (r2 = 0.853··); Alcohol & Anthocyanis (r2 = 0.862··); Total acidity & Volatile acidity (r2 = 0.601··); Total acidity & Non-Reducible extract (r2 = 0.715··); Total acidity & Coloring intensity (r2 = 0.659··); Total acidity & Color tint (r2 = 0.611··); Total acidity & Total polyphenols (r2 = 0.722··); Total acidity & Anthocyanis (r2 = 0.759··); Volatile acidity & Non-Reducible extract (r2 = 0.671··); Volatile acidity & Coloring intensity (r2 = 0.724··); Volatile acidity & Color tint (r2 = 0.725··); Volatile acidity & Total polyphenols (r2 = 0.706··); Volatile acidity & Anthocyanis (r2 = 0.691··); Non-Reducible extract & Sugar content (r2 = 0.691··); Non-Reducible extract & Coloring intensity (r2 = 0.877··); NonReducible extract & Color tint (r2 = 0.855··); Non-Reducible extract & Total polyphenols (r2 = 0.881··); Non-Reducible extract & Anthocyanis (r2 = 0.917··); Coloring intensity & Color tint (r2 = 0.996··); Coloring intensity & Total polyphenols (r2 = 0.944··); Coloring intensity & Anthocyanis (r2 = 0.948··); Color tint & Total polyphenols (r2 = 0.926··); Color tint & Anthocyanis (r2 = 0.930··); Total polyphenols & Anthocyanis (r2 = 0.952··) (Table 4). Regarding negative correlations it can be observed that in all the analyzed cased there was a weak negative correlation Alcohol & Free SO2 (r2 = -0.336·); Volatile acidity & Free SO2 (r2 = -0.301·); Volatile acidity & Total SO2 (r2 = -0.459·) (Table 4).

Concerning Pearson correlation coefficient between metals from wine (Table 5) there are small number of both positive and negative correlations between the metal concentrations of the analysed wines. There are some relevant examples: K & Na (r2 = 0.252·); K & Ca (r2 = -0.411·); K & Li (r2 = 0.247·); Na & Mg (r2 = 0.238·); Mg & Fe (r2 = -0.501··); Fe & Mn (r2 = -0.407··); Fe & Zn (r2 = - 0.393··); Cu & Zn (r2 = -0.272·); Mn & Zn (r2 = 0.250·).

Based on the previous Pearson correlation index, through this present research have been shown that the main parameters analysed from wine have had an influence on each other; in other words, the quality of the wine produced in the Vineyard of Dealu Bujorului is directly contingent on all these parameters.

Combining the physico-chemical characteristics of red wines with elemental concentration for wine geographical discrimination

Multivariate chemometric method was applied for the differentiation of wines intro groups on the basis of their geographic origin. Stepwise linear discriminant analysis (LDA) was used to identify significant tracers for classification to the geographical discrimination of the wines samples. By cross-validation, we established the optimal number of parameters required to obtain a robust model.

Based on the physico-chemical characteristics and elemental concentration the cross-validation technique provided a 75.31% percentage of predicted membership according to the wine geographic origin (F1 = 52.67% and F2 = 22.64%) (Figure 1). A significant differentiation of wines according to physico-chemical characteristics and elemental concentration was carried out for wines samples, which demonstrates the importance of the physico-chemical characteristics and elemental concentration for the geographical traceability of wines.

The differentiation of wines according to geographic origin based on the elemental concentration of wine, in this case a 70.63% percentage of predicted membership according to the wine geographic origin (F1 = 39.21% and F2 = 31.44%) (Figure 2). The differentiation of wines according to geographic origin based on the physico-chemical characteristics of red wine, in this case a 85.15% percentage of predicted membership according to the wine geographic origin (F1 = 63.66% and F2 = 21.50%) (Figure 3).

Conclusions

Based on the results regarding the qualitative assessment of the tested varieties, they have a very good suitability in the studied areas. In terms of quality rating, they display particular characters of the varieties, as well as the ecoclimatic conditions and ecopedological influence on the quality of wine.

This higher content of some metals may be due to the use of fertilizers for cultivation of vine (K, Ca, Cu) the winemaking process or addition of substances for wine clearing such as bentonite (Na, Ca, Fe) or alos, and the viticultural practices. Cu content is below the limit of detection due to the modern technology for obtaining wines in a controlled manner. The work offers new information on the quality of the white wines obtained in main vineyards of Romania, useful for their promotion and marketing.

Based on the previous Pearson correlation index, through this present research have been shown that the main parameters analysed from wine have had an influence on each other; in other words, the quality of the wine produced in the Vineyard of Dealu Bujorului is directly contingent on all these parameters.

Based on the physico-chemical characteristics and elemental concentration, a relevant discrimination of wines according to their geographical origin and years was performed. The variation of the physico-chemical characteristics and elemental concentration represents a strong geological marker for wines geographical traceability. The proposed methodology allowed an 85.15% successful classification of wines according to the region of provenance and also the years of wine obtaining.

Acknowledgments

This paper was published under the frame of the Romanian Ministry of Agriculture and Rural Development, project ADER no. 14.2.2. "Quantitative studies on assessment and monitoring contaminants, on the chain of viticulture and winemaking to minimize the amount of pesticides and heavy metals as principal pollutants".

References

References

Artimon, M., Dumitrescu, V., Tănase, I.G., Pele, M. 2008. The optimization of the methods for Cu, Zn and Pb content determination in Romanian wines by AAS afther dry or microwave mineralization. Romanian Biotechnological Letters, 14(2), 4319-4325.

Avram, V., Voica, C., Hosu, A., Cimpoiu, C., Măruţoiu, C. 2014. ICP-MS characteziation of some Romanian white wines by theier mineral content. Revue Roumaine de Chimie, 59(11-12), 1009-1019.

Álvarez-Iglesias, P., Rubio, B., Millos, J. 2012. Isotopic identification of natural vs. anthropogenic lead sources in marine sediments from the inner Ría de Vigo (NW Spain). Science of the Total Environment, 437, 22-35.

Baiano, A., De Gianni, A., Previtali, M. A., Del Nobile, M.A. 2015. Effects of defoliation on quality attributes of Nero di Troia (Vitis vinifera L.) grape and wine. Food Research International, 75, 260-269.

Bora, F.D., Pop, T.I., Babeş A.C., Popescu, D., Iliescu, M., Pop, N. 2015a. Research on the quality of the three white wines varieties in Transylvania, harvest of 2013-2014. Bulletin UASVMHorticulture, 72(2), 327-334.

Bora, F.D., Bunea, C.I., Rusu, T., Pop, N. 2015b. Vertical distribution and analysis of micro-, macroelements and heavy metals in the system soil-grapevine-wine in vineyard from North-West Romania. Chemistry Central Journal, 9(19), 1-13.

Bora, F.D., Donici, A., Ciubuca, A., Postolache, E., Tabaranu, G., Enache, V., Bîrliga, N., Pop, N., Bunea, C.I. 2016a. Qualitative assessment of the red wine varieties grown in Dealu Bujorului vineyard. Bulletin UASVM Horticulture, 73(2), 116-125.

Bora, F.D., Donici, A., Oşlobanu, A., Fiţiu, A., Babeş, A.C., Bunea, C.I. 2016b. Quality assessesment of the white wine varieties grown in Dealu Bujorului vineyard, Romania. Notulae Botanicae Horti Agrobotanici, 44(2), 593-602.

Bora, F.D., Donici, A., Rusu, T., Bunea, A., Popescu, D., Bunea C.I. 2018. Elemental profile and 207Pb/206Pb, 208Pb/206Pb, 204Pb/206Pb, 87Sr/86Sr Isotope ratio as fingerprints for geographical traceability of Romanian wines. Notulae Botanicae Horti Agrobotanici, 46(1), 223-239.

Budić-Leto, I., Mešin, N., Gajdoš, K., Pezo, I., Bralić, M., 2009. Comparative study of the total acidity determination in wine by potentiometric and volumetric titration. Agriculturae Conspectus Scientificus, 72(1), 61-65.

Bruijn, J., Loyola C., Arumí L.J., Martínez J. 2014. Effect of non-protein factors on heat stability of Chilean Sauvignon Blanc wines. Chilean Journal of Agricultural Research, 74(4), 490-496.

Catarino, S., Madeira, M., Monteiro, F., Caldeira, I., Rosa, T., Martins, P., Bruno de Sousa, R., Curvelo-Garcia, A. 2014. Multi-elemental analysis throughout soil-wine system as a generator of information on geographic origin. Conference: 37th World Congress of Vine and Wine, At Mendoza, Argentina.

Coetzee, P.P., van Jaarsveld, F.P, Vanhaecke, F. 2014. Intraregional classification of wine via ICP-MS elemental fingerprinting. Food Chemistry, 164, 485-492.

Di Paola-Naranjo, R.D., Baroni, M.V., Podio, N.S., Rubinstein, H.R., Fabani, M.P, Badini, R.G, Inga, M., Ostera, H.A., Cagnoni, M., Gallegos, M., Gallegos, E., Gutier, E., PeralGarcia, P., Hoogewerff, J., Wunderlin, D.A. 2011. Fingerprints for main varieties of Argentinean Wines: Terroir differentiation by inorganic, organic, and stable isotopic analyses Coupled to Chemometrics. Journal of Agricultural and Food Chemistry, 59, 7854-7865.

Dos Santos, C.E.I., da Silva, L.R.M., Boufleur, L.A., Debastini, R., Stefano, A.C., Amaral, L., Yoneama, M.L., Diaz, J.F. 2010. Elemental characterisation of Cabernet Sauvignon wines using particule-induced X-ray Emission (PIXE). Food Chemistry, 121, 244-250.

Ðurðić, S., Pantelić, M., Trifković, J., Vukojević, V., Natić, M., Tešić, Z., Jutić, J. 2017. 2017. Elemental composition as a tool for the assessment of type, seasonal variability, and geographical origin of wine and its contribution to daily element intake. RSC Advances, 7, 2151-2162.

Fabani, M.P., Arrúa, R.C., Vázquez, F., Diaz, M.P., Baroni, M.V., Wunderlin, D.A. 2010. Evaluation of elemental profile coupled to chemometrics to assess the geographical origin of Argentinean wines. Food Chemistry, 119, 372-379.

Geana, E.I., Sandru, C., Stanciu, V., Ionete, R.E. 2016. Elemental profile and 87Sr/86Sr isotope ratio as fingerprints for geographical traceability of wines: an approach of Romanian wine. Food Analytical Methods, 10(1), 63-73.

Geana, E.I., Ionete, R.I., Tudorache, A., Pasa, A, Postolache, E., Ranca A. 2011. Phenolic contentes of Romanian wines with different geographic origins. Asian Journal of Chemistry, 23(12), 5197-5201.

Geana, E.I., Marinescu, A., Iordache, A.M., Sandru, C., Ionete, R.I., Bala, C. 2014. Differentiation of Romanian wines on geographical origin oand wine variety by elemental composition and phenolic components. Food Analytical Methods, 7(10), 2064-2074.

Iglesias, M., Besalú, E., Anticó, E. 2007. Internal standardization-atomic spectrometry and geographical pattern recognition techniques for the multielement analysis and classification of Catalonian red wines. Journal of Agricultural and Food Chemistry, 55(2), 219-225.

Ivanova-Petropulos, V., Wiltsche, H., Stafilov, T., Stefova, M., Motter, H., Lankmayr, E. 2013. Multi-element analysis of Macedonian wines by inductively coupled plasmamass spectrometry (ICP-MS) and inductively coupled plasma-optical emission spectrometry (ICP-OES) for their classification. Macedonian Journal of Chemistry and Chemical Engineering, 32(2), 265-281.

Jackson, D.I., Lombard, PB. 1993. Environmental and management practices affecting grape composition and wine quality-a review. American Journal of Enology and Viticulture, 44, 409-430.

Karatas, D., Aydin, F., Aydin, I., Karatas H. 2015. Elemental composition of red wines in Southeast Turkey. Czech Journal of Food Sciences, 33(3), 228-236.

Kment, P., Mihaljević, M., Ettler, V., Šebek, O., Strnad, L., Rohlová, L. 2005. Differentiation of Czech wines using multielement composition-A comparison with vineyard soil. Food Chemistry, 91(1), 157-165.

Lara, R., Cerutti, S., Salonia, J.A., Olsina, R.A., Martinez, L.D. 2005. Trace element determination of Argentine wines using ETAAS and USN-ICP-OES. Food Chemistry Toxicol, 43(2), 293-297.

Miličević, B., Šubarić, D., Jozinović, A., Ačkar, Ð., Babić, J., Vuković, D., Mrgan, A., 2014. Effect of the immobilized yeast cells fermentation on chemical composition and biogenic amines content in wine. Food Technology and Biotechnology, 15, 276-282.

Mitic, M.N., Obradovic, M.V., Grahovac Z.B., Pavlovic A.N. 2010. Antioxidant capacities and phenolic levels of different varieties of Servian white wines. Molecules, 15(3), 2016-2027.

Pohl, P. 2007. What do metals tell us about wine? TrAC Trends in Analytical Chemistry, 26(9), 941-949.

Postolache, E., Ciubuca, A., Bunea, C.I., Moldovan, M.P., Pop, N., Donici, A., Bora, F.D. 2016. Research on quality of red wine varieties obtained at Bujoru vineyard, Dealu Bujorului Wine Center, Romania. Advance in Agriculture & Botanics - Bioflux, 8(2), 58-76.

Seguin, G. 1986. 'Terroirs' and pedology of wine growing. Experientia, 42(8), 861-873.

Ražić, S., Onjia, A. 2010. Trace element analysis and pattern recognition techniques in classification of wine from Central Balkan Countries. American Journal of Enology and Viticulture, 61(4), 506-511.

Rodrigues, S.M, Otero, M, Alves, A.A, Coimbra, J, Coimbra, M.A, Pereira, E, Duarte, A.C 2011. Elemental analysis for categorization of wines and authentication of their certified brand of origin. Journal of Food Composition and Analysis, 24(4-5), 548-562.

Trigo-Córdoba, E., Bouzas-Cid, Y., Orriols-Fernández, I., Mirás-Avalos, J.M., 2015 Effects of deficit irrigation on the performance of grapevine (Vitis vinifera L.) cv. 'Godelio' and 'Treixadura' in Ribeiro, NW Spain. Agricultural Water Management, 161, 20-30.

Ţârdea, C., Sârbu, H., Ţârdea, A. 2001. Wine treated. Publisher „Ion Ionescu de la Brad", Iasi, Romania.

Vrček, I.V., Bojić, M., Žuntar, I., Mendaš, G., Medić-Šaric, M. 2011. Phenol content, antioxidant activity and metal composition of Croatian wines deriving from organically and conventionally grown grapes. Food Chemistry, 124(1), 354-361.