Effect of Vineyard Location on Assyrtiko Grape Ripening in Santorini and Its Wine’s Characteristics ^†

Abstract

Besides the other factors, the microclimate (terroir) influences the quality characteristics of wine. The Assyrtiko variety has adapted to the volcanic soil of Santorini but under climate change, finding an ideal location for full grape ripening represents a challenge in preserving the PDO quality of Santorini wines. Thus, this study aims to evaluate the effect of location and harvesting time on the quality of Assyrtiko wine. It assessed the location effect (three distant plots of land in three distant areas of the island) on the composition of grapes (water uptake, pH, sugar, and organic acid accumulation) during the ripening. The grapes were vinified using the same procedure. The aromatic volatile profile of the wines was evaluated with GC-MS. A two-way ANOVA revealed that besides location and harvest time, their interaction is also significant for the parameters studied, except for the interaction effect involving sugar content. The analysis of volatile aromatic compounds revealed that the wine from grapes harvested at a later date had a higher aromatic intensity with notes of citrus, white-fleshed, and tropical fruits. This wine surpasses the levels of 2-phenylethanol, isoamyl acetate, linalool, and 2-phenylethyl ester with 17.8%, 7.7%, 21.1%, and 15.6%, respectively, compared to the immediate next in descending order. Results suggest that the grape variety is better suited to the local climatic conditions when full grape ripeness is reached by the end of the growing season.

1. Introduction

Earlier harvesting is associated with low sugar concentration and incomplete ripening. Both nowadays are frequent phenomena in viticultural zones around the world because of globally increasing temperatures associated with climate change. Wine grape (Vitis vinifera L.) vineyards cover more than 7.2 million hectares worldwide, with Greek vineyards representing 0.9% of this area according to [1]. In 2023, the estimated net worth of the wine industry was more than EUR 333 billion [2]. There are more than 300 Greek indigenous varieties cultivated as single or in combination with international varieties in nine different wine-growing regions (mainland and islands) in soils that vary strongly in altitude (from 0 to over 1000 m). Greece has 33 Protected Designation of Origin (PDO) zones and more than 100 Protected Geographical Indication (PGI) zones. Although the international wine trade and consumers had specific demands for international varieties, especially French ones, and the wine grape varietal choices narrowed across the world and became less diversified, nowadays, winegrowers consciously select the cultivar of wine grapes based on the phenotypic traits that best match their (micro-)climates and soils [3]. The best expression of terroir is achieved when the precocity of the grapevine variety is suited to the local climatic conditions so that full ripeness is reached by the end of the growing season [4].

Over the last 500 years, but particularly the last four decades, regional warming and drying climatic conditions have accelerated in the Mediterranean basin [5,6]. Compared with the mainland, islands have less adaptive capacity, being more vulnerable to the impacts of climate change. The new conditions, having less access to water and energy resources, directly affect the local economy as their agricultural production and tourism services are directly affected [7].

The literature reports that, in general, water deficit accelerates sugar accumulation and malic acid breakdown in grape juice, while early water deficit during the growth period increases phenolics in berry skins; the resulting wines have higher levels of bound volatile compounds and have higher scores in tasting trials [8]. In addition, in conditions of controlled grapevine clone and winemaking protocols, both soil and microclimate conditions that characterize a vineyard significantly affect wine composition [9]. The vineyard’s location affects amino acids in grape juice and, consequently, fermentation and final wine composition [10], but it also strongly affects terpene content in wine [11]. Vineyards show significant soil variability not only between regions but also between plots. Minerals in the soil, such as N, P, and K, have an important impact on grape berry development, alcoholic fermentation, and wine quality [12,13], suggesting the need for the optimization of nutrients to attain optimal grape quality and avoid the production of imperfect wines [14,15]. Yet, soil health is linked with microclimate parameters such as temperature and rainfall; it increases when temperature drops and there is a rise in rainfall rates and soil clay content [16].

Santorini, located in the Aegean Sea, is a volcanic Greek island with unique morphological peculiarities and characteristics and a long history of grape cultivation and wine production; it is one of the main sources of agricultural production [17]. Successive volcanic eruptions, the largest of which took place around 1600 BC, were a milestone, among other things, for the development of viticulture and wine production on the island [18]. The peculiar local ecosystem created by the successive eruptions of the local volcano favors indigenous grape varieties, with low productivity and high quality of the wine produced. Santorini remains one of the few phylloxera-free wine-growing areas, where vineyards over 100 years old exist. The viticultural region of the island comprises approximately 1000 ha from sea level to the terraces at 150 to 350 m above sea level. There are two specific and unique training systems which are well-adapted to the island’s specific climate and weather conditions, namely the “kouloura” (basket shape) and the “kladeftiko” [19]. Assyrtiko is a rare ancient grape variety of Greece and is considered a Greek wine ambassador. This variety, perfectly adapted to the unique climate of the volcanic island of Santorini, is arguably the most intriguing Greek white cultivar [20]. It is the first Greek grape variety planted for commercial reasons in other countries and continents, and according to the specifications for the controlled designation of origin “Languedoc” in France since 2024, it participates as an auxiliary variety in a percentage equal to or less than 5% of its varietal composition [21].

Since 1971, the exquisite quality of Santorini wines has been certified by the EU’s Protected Designation of Origin, which defines and safeguards its particular characteristics exclusive to the island’s geographical environment and inherent qualities. In 2018, Santorini was inserted in the National Register of Intangible Cultural Heritage regarding its wine traditions and the particular way of cultivating vineyards, with this development being the first step towards inclusion in the UNESCO list. The world-famous Santorini wines are facing unprecedented challenges, as their crops are being affected by drought and the shrinking of the local viticulture terroir, a result of the unregulated touristic development of land on the island [22].

Xyrafis, Alain, Petoumenou, Ioannis, and Biniari [19] analyzed the climate data of Santorini over time, evaluating trends in climate parameters and bioclimatic indices and comparing them to viticultural indices. They found that in the region, the average annual temperature has risen by over 4 °C over the past 45 years. Additionally, there has been a significant increase in the frequency of hot days. Bioclimatic indicators indicate warmer climates, longer drought spells, and warmer nights. High temperatures during vine development and differentiation might impact productivity in the next season, resulting in early harvest dates and higher sugar levels.

The Assyrtiko wine grape variety, well adapted to adverse climatic conditions, represents an example of growing techniques and traits to be chosen [19]. The majority of research on Assyrtiko grapes in Santorini has been concentrated on viticultural methods (sprays of kaolin and CaCO₃) [23] and grape microbiota during grape maturity [24]. As mentioned previously, besides vineyard management, the grape and wine quality is greatly influenced by environmental factors, soil composition, and microclimate. Papas, Ftelos, and Louros are well-known grape areas in Santorini, with vineyards dating more than a century ago. Nevertheless, to our knowledge, no scientific study exists regarding the quality characteristics of the grapes from these locations and the resulting wines. Addressing this research gap could provide valuable insights into how distinct vineyard locations within Santorini influence Assyrtiko grape ripening and wine characteristics, thereby informing viticultural practices and enhancing wine quality.

In this context, we aimed to study the effect of the Santorini vineyard location on the Assyrtiko ripening process and its impact on the quality characteristics of the produced wines. In order to fulfill this aim, three locations (Louros, Papas, and Ftelos) that are reported to represent different expressions of the Assyrtiko variety were chosen. The winemaking procedure was conducted at a local winery in Santorini. Grapes from each selected vineyard were harvested manually and vinified individually at the winery.

2. Materials and Methods

2.1. Grape Sampling and Harvest

Sampling (a minimum of two samples per vineyard) began on 16 August 2022 at the end of the veraison. Three vineyards, non-irrigated, in different altitudes and parts of the island, where the Assyrtiko winegrape variety has been growing for over 50 years, were chosen for this research; these are Louros (36.382353, 25.448229), Ftelos (36.416128, 25.433972), and Papas (36.376470, 25.433113). Sampling was carried out on fixed days early in the morning, with a random collection of 300 grapes (any size, any stage of maturity) into sterile plastic bags. Samples were transferred directly to the winery’s oenological laboratory to monitor the growth of the grape berries. The analyses of potential alcohol content, pH value, and total acidity of the grapes, as well as alcoholic degree, total acidity, pH value, volatile acidity, and residual sugar content of the produced wines were determined according to the methods of the International Organization of Wine and Vine [1]. Additionally, total solid residue (OIV-MA-AS2-03B), malic acid, tartaric acid, catechins, and assimilable nitrogen were determined with an enzymatic analyzer Hyperlab (Steroglass, Italy).

2.2. Vinification

Winemaking was performed in a local winery of Santorini (Artemis Karamolegos, Exo Gonia, Santorini, Greece). Grapes from each selected vineyard were harvested separately, manually, and upon arrival, were placed for 12 h into the winery’s refrigerator. Rotten stems and grapes were removed by hand on a sorting table, followed by destemming and crushing (Bucher, Berlin, Germany). Then, grapes were transferred to a pneumatic press (Bucher, Germany), where they remained for 5–6 h with dry ice (average temperature 8–10 °C) and liquid sulfite was gradually added (100 mL/hL), as well as oenological tannin (Vitanil B/Martin Vialatte, Magenta, France) (100 gr/hL). At the extraction step, pectinolytic enzymes (c-max/Lallemand) (2 gr/hL) were added. After static settling (12 h at 12 °C), clear must was transferred by a peristaltic pump (Bucher, Germany) in a 1000 L stainless steel tank. Each batch was inoculated with 30 gr/hL of active dry yeast Saccharomyces cerevisiae (commercial name cross-evolution Lallemand, Blagnac, France) and before inoculation, the following were added: glutathione (22 gr/hL) (Glutastar), nutrient organic azote FermaidO (40 gr/hL) (Lallemand), tartaric acid (127 gr/hL), and DAP FermaidE (Lallemand) (50 gr/hL) combined with oxygenation for 5 min/tn. Fermentations were carried out between 16 and 18 °C.

2.3. Gas Chromatography–Mass Spectrometry (GC-MS/MS)

Volatiles were isolated using the following extraction procedure: wine 3 mL, water 7 mL, ammonium sulfate 4.5 g, dichloromethane 1.5 mL, and phenol (as internal standard) 10 µL (10 mg/L) were added to a glass test tube. First, the mixture was vortexed for 60 s and then centrifuged for 4 min at 4000 rpm. The upper phase was discarded, and 0.5 mL of the organic phase was transferred to a vial containing 0.4 g of sodium sulfate. A volume of 1 µL of the final solution was injected into the GC-MS/MS instruments.

GC-MS analysis was performed using a Quantum XLS Gas Chromatograph (Thermo Scientific Inc., Waltham, MA, USA) coupled to a Triple Quad Mass Spectrometer (Thermo Scientific Inc., Waltham, MA, USA). The column, TR-Pesticide II (30 m × 0.25 mm ID, 0.25 µm film thickness, Thermo Scientific) with helium as the carrier gas (1.0 mL/min) and the temperature program employed had an initial temperature of 40 °C for 5 min, increased by 7 °C/min to 170 °C, then increased by 40 °C/min to 290 °C and held for 3 min. The injection volume was 2 µL in the splitless mode, with an injection inlet temperature of 210 °C.

MS/MS settings were as follows: experiment type SRM (selective reaction monitoring), collision gas pressure 1.5 mTorr, Q1 and Q3 peak width 0.70, cycle time 0.500 (s), solvent delay of 7 min, emission current 50 µA, source temperature 200 °C, transfer line temperature 250 °C, and each volatile compound was injected into the gas chromatograph at a concentration of 10 mg/L. The experimental mass spectra were compared to the mass spectra database from NIST20. More information regarding calibration, MS parameters, and the method validation are reported elsewhere [25].

2.4. Sensory Evaluation—Questionnaire

The sensory evaluation panel consisted of 10 oenologists (five women and five men) working in wine production, quality control, and consulting. The three wine samples were primarily coded and blind-tasted. The common origin of the wines evaluated optimizes the reliability of the results. It minimizes possible errors due to the differences in the climatic (temperature, rainfall, atmospheric humidity) or soil conditions and grape growth. Winetasters were asked to rate wines’ questioned characteristics on a scale from 1 to 5 (1 being the least, 5 the most). The questions examined the following parameters of wine: color intensity, aroma description, aromatic intensity, and the taste of each wine (acidity, roundness, complexity, flavor balance, aftertaste). The questionnaire and the study received approval from the Ethics Committee of the Democritus University of Thrace.

2.5. Statistical Analysis

The statistical data analysis was performed using a one-way ANOVA followed by Duncan’s post hoc test. A two-way ANOVA was applied to understand if two independent variables (location and sampling date), in combination, affected the responses. A p-value of 0.05 was considered for all the statistical tests performed. The data were statistically evaluated with SPSS Statistics (software version: 26, Chicago, IL, USA).

3. Results

3.1. Soil–Climatic System and Ripening

The 2021–2022 wine season had mild weather and rainfall that exceeded the last decade’s average, and the harvest was characterized as a “particularly late” one. The recorded precipitation between October 2021 and mid-March 2022 was 320 mm, almost 70% higher than the same period of the previous year. Spring was warm without typical strong winds, which is often the cause of significant damage to the vines during the sensitive period of growth. According to Leeuwen et al. [26], vine water status is a key determinant of terroir expression, which depends on climatic conditions such as rainfall and evapotranspiration. All the above contributed to a slow and smooth ripening. As a result, grapes were produced with a high level of acidity, as well as high-quality grapes. During the harvest period, the absence of a heat wave resulted in higher yields than the 2021 harvest. Ripening began gradually, starting with the vineyards at the lowest altitudes of the island. There was an interruption of the dry summer weather with heavy rainfall (27 mm) on August 24 (

Table 1. Grape physicochemical characteristics during the ripening period *^,**.

Vineyard	Sampling Date	100-Grape Weight (g)	Total Acidity (g Tartaric Acid/L)	pH	Sugar (Brix)	Malic Acid (g/L)	Tartaric Acid (g/L)	Catechins (mg/L)	Yeast Assimilable Nitrogen (mg/L)
Ftelos	16 August 2022	226 1 ± 2.00	7.88 2 ± 0.03	3.01 1 ± 0.01	20.10 1 ± 0.05	1.38 2 ± 0.01	10.85 2 ± 0.05	12.00 2 ± 1.00	129.35 2 ± 0.35
	22 August 2022	234 2 ± 1.80	7.70 1 ± 0.10	2.99 1 ± 0.01	21.90 2 ± 0.03	1.03 1 ± 0.02	8.76 1 ± 0.04	7.00 1 ± 0.01	87.50 1 ± 0.28
Louros	16 August 2022	219 2 ± 1.00	7.47 3 ± 0.06	2.94 1 ± 0.01	19.90 1 ± 0.04	1.16 4 ± 0.02	10.02 4 ± 0.03	12.00 2 ± 0.01	166.96 4 ± 0.26
	21 August 2022	232 3 ± 0.87	7.40 3 ± 0.05	3.00 2 ± 0.01	21.83 2 ± 0.06	0.84 3 ± 0.01	7.36 1 ± 0.04	7.00 1 ± 1.00	73.57 3 ± 0.17
	25 August 2022	240 4 ± 2.65	6.60 2 ± 0.05	3.13 3 ± 0.02	21.80 2 ± 0.04	0.74 2 ± 0.01	9.16 3 ± 0.06	7.00 1 ± 0.01	57.26 2 ± 0.14
	31 August 2022	215 1 ± 1.00	6.15 1 ± 0.05	3.10 3 ± 0.05	22.20 3 ± 0.03	0.58 1 ± 0.01	8.82 2 ± 0.02	7.00 1 ± 0.01	55.42 1 ± 0.17
Papas	16 August 2022	209 1 ± 0.01	7.20 1 ± 0.10	2.95 1 ± 0.01	20.00 1 ± 0.20	1.22 2 ± 0.02	10.00 2 ± 0.05	11.00 2 ± 1.00	192.80 2 ± 0.15
	21 August 2022	230 2 ± 1.00	7.60 2 ± 0.05	3.09 2 ± 0.01	21.70 2 ± 0.06	0.78 1 ± 0.01	7.92 1 ± 0.61	6.00 1 ± 0.01	106.13 1 ± 0.14

* Data are reported as mean ± standard deviation. ** Different numbers for the same location are significantly different (p < 0.05) by Duncan’s multiple range test.

). This heavy rain delayed the harvesting and caused a decrease in acidity and sugars, but with delayed time, they reached the desired level in the fruit.

A two-way ANOVA examined the effect of harvesting time, location, and their interaction on the four parameters. Besides the main effects, the interaction between harvesting time and location is significant for the four parameters studied (except for interaction on the Brix level of must). As expected with the ripening, there is an increase in the weight of the grapes (correlation coefficient 0.423, p = 0.039), pH (0.774, p = 0.014), and Brix (0.953, p = 0.000) values. On the other hand, TA (−0.953, p = 0.000) and concentrations of malic, tartaric, and yeast assimilable nitrogen significantly decrease with ripening.

The profile of the vineyard plots is reported in

Table 2. The profile of the vineyard plots studied.

Vineyard	Altitude (m)	pH	Organic Substance (%)	Clay (%)	Phosphorus (ppm)	Potassium (ppm)
Louros	250	6.96 ± 0.02	0.29 ± 0.01	8.10 ± 0.50	34.42 ± 1.05	141.71 ± 1.05
Ftelos	200	7.04 ± 0.03	0.65 ± 0.02	7.29 ± 0.45	4.05 ± 0.30	232.82 ± 1.30
Papas	290	6.83 ± 0.02	0.47 ± 0.02	7.80 ± 0.47	29.35 ± 0.80	185.76 ± 1.10

Data are reported as mean ± standard deviation.

. The soil composition of the three vineyards studied is sandy clay, with Papas having a higher percentage (84%), while the other two are almost the same (68–70%). The organic substance is, in all cases, less than 1%. The pH is neutral for grape cultivation and ranges from 6.5 to 7.5, i.e., neutral to weakly alkaline. There is a significant difference in phosphorus and potassium content among the locations, with Ftelos being lowest in P and highest in K. The opposite was observed in Louros.

3.2. Alcoholic Fermentation and Characteristics of Wine

The fermentations were carried out at the same range of temperatures (16–17 °C). Although the exponential growth phase in the three wines has the same duration, the fermentation lasted 24 days for Ftelos, 23 days for Louros, and 31 days for Papas. The variation in the residual sugars of each wine upon the completion of alcoholic fermentation is reported in Figure 1.

The chemical characteristics of wines are reported in

Table 3. Chemical characteristics of the wines at the end of alcoholic fermentation.

	Residual Sugars (g/L)	Alcoholic Strength by Vol (%)	Total Acidity (g/L Tartaric Acid)	pH	Volatile Acidity (g/L Acetic Acid)	Density 20 °C (g/L)	Total Solid Residue (g/L)
Louros	0.49 ± 0.03	15.00 ± 0.05	7.88 ± 0.15	2.89 ± 0.05	0.56 ± 0.03	0.9887 ± 0.0001	24.40 ± 0.05
Ftelos	1.21 ± 0.10	14.20 ± 0.10	8.07 ± 0.15	2.90 ± 0.03	0.41 ± 0.04	0.9896 ± 0.0002	23.80 ± 0.03
Papas	0.12 ± 0.02	14.00 ± 0.08	7.57 ± 0.10	2.94 ± 0.05	0.44 ± 0.03	0.9890 ± 0.0001	24.90 ± 0.05

Data are reported as mean ± standard deviation.

. The wines produced had more than 14.00% alcoholic strength, with a significant difference between the Louros wine and the other two. This is because the grapes of Louros’ vineyard were harvested later than the other two. Tartaric acid was added to all wines depending on the potential alcoholic strength to reach a final acidity of 6.2–6.5. Wine from the Ftelos vineyard has the highest total acidity, which is not accompanied by the lowest pH potentially because of the production of some weak acids during fermentation. The wine originating from the Papas vineyard has the lowest acidity and the highest pH. The high values of total acidity and low pH of the final wines are within the range usually found in Assyrtiko wines [27]. All pH values of the produced wines were in a very close range of 2.89–2.94, with a difference occurring only between Louros and Papas wines, which, however, due to the close values and a maximum difference of 0.05, is not considered significant for differentiation in the quality of the wine. The measurement of volatile acidity expressed in the concentration (g/L) of acetic acid in the final wines showed the lowest value of 0.41 g/L in Ftelos, without having a big difference with Papas’s wine. The highest volatile acidity was observed in the Louros wine (0.56 g/L), which completed faster alcoholic fermentation than the other two wines. The increased volatile acidity may be due to the contact of the wine with oxygen in the early stages of alcoholic fermentation or during the pressing of the grapes.

According to Spearman’s rho correlation, altitude negatively affects the residual sugars (−0.957, p = 0.000) and TA (−0.934, p = 0.000) of wines but the must pH positively (0.667, p = 0.050). That is, the higher the altitude, the higher the pH value and the later the harvest [28]. On the other hand, the weight of grapes is negatively correlated with ethanol content (−0.896, p = 0.001). Obviously, the decrease in berry weight, probably due to dehydration, causes an increase in grape sugar concentration and consequently increases the wine alcohol concentration.

3.3. Volatile Profile of Wines

Gas chromatography combined with spectrometry was performed to assess the volatile profile of wines (

Table 4. Qualitative and quantitative results of gas chromatography combined with mass spectroscopy GC-MS/MS.

Volatile Compound	Ftelos (μg/L)	Louros (μg/L)	Papas (μg/L)	Threshold (μg/L)	Aroma Description
2-phenyl ethanol	30,000	53,000	45,000	620	Floral, rose with a hint of honey
Guaiacol	<10 *	<10 *	<10 *	9.5	Phenol, wood
2-methoxy-4methylphenol	<10 *	<10 *	<10 *		Bacon, clove
Hexanol	<10 *	<10 *	<10 *	8.000	Green, grassy
Ethyl-2-methylbutyrate	<10 *	<10 *	<10 *	18	Strawberry, apple
Ethyl phenol	<10 *	<10 *	<10 *	440	Spices, cloves, tobacco
Ethyl butyrate	<10 *	<10 *	<10 *	125	Cherry, strawberry, apple, pineapple, sweet
Ethyl caprylate	289	266	247	150	Pear, apple, pineapple
Ethyl caproate	975	885	886	100	Pineapple, apple peel
Ethyl cinnamate	<10 *	<10 *	<10 *	1.1	Flowers, red fruits
Ethyl dodecanoate	<10 *	<10 *	<10 *	1750	Plum, grape
Ethyl decanoate	109	117	103	200	Red fruits, grapes
Ethyl isobutyrate	127	112	163	10	Sweet scent, strawberry, floral
Ethyl isovalerate	<10 *	<10 *	<10 *	3	Yeast, fruit
Ethyl-3-hydroxybutyrate	979	881	873	21,000	Fruity, green, apple peel
2-phenylethyl acetate	431	505	437	250	Flower, rose
Isoamyl acetate	7986	8599	6533	30	Banana, candy
Isobutyl acetate	206	208	196	1.600	Fruit, apple, banana
Hexyl-acetate	557	475	512	670	Apple, cherry, peach, flower
Benzyl acetate	<10 *	<10 *	<10 *		Jasmine, apple
Eugenol	<10 *	<10 *	<10 *	6	Clove
Thymol	<10 *	<10 *	<10 *	50	Thyme, oregano
Geraniol	<10 *	<10 *	<10 *	30	Rose, citrus
Linaool	159	241	199	25	Lemon, citron, bergamot
B-ionone	<10 *	<10 *	<10 *	4.3	Violet, plum, raspberry
Citral	<10 *	<10 *	<10 *		Lemon
Damasceone	<10 *	<10 *	<10 *	0.05	Cooked apple, quince, flowers
Rose oxide	<10 *	<10 *	<10 *	80–160	Rose, lychee
Decyl-aldehyde	<10 *	<10 *	<10 *		Orange peel, citrus fruit, flower
Ethyl vanillin	<10 *	<10 *	<10 *	200	Flower fragrance
Isoeugenol	<10 *	<10 *	<10 *		Flower, aromatic vanilla

* Concentration below the limit of quantification (LOQ).

). According to our results, the major components of Assyrtiko wines from the Santorini PDO zone were, in reducing order, 2-phenylethanol, isoamyl acetate, ethyl caproate, ethyl-3-hydroxybutyrate, hexyl- acetate, 2-phenethyl acetate, isobutyl acetate, and linalool. The same aromatic compounds were detected in all wines but in various concentrations. Previous studies [20,29] have reported the aromatic profile of Assyrtiko wines from Santorini. Their results were comparable regardless of being focused on the effect of pre-fermentative treatment [20] or nitrogen availability and the S. cerevisiae strain [29].

The literature suggests that different chemical components contribute to wine aroma. Moreover, their contribution depends on the quantity in which they are present relative to their respective threshold [30,31]. Higher alcohols and esters are produced depending on the fermentation conditions and yeast strain. The same yeast strain, known to create wines with aromatic intensity (notes of fresh fruit and floral smells) and ester formation, was used to ferment all our wines. A member of the higher alcohol group, 2-phenylethanol, which smells like flowers, roses, and honey and is frequently found in wine and beer at concentrations between 20 and 180 mg/L, was detected in all our samples in high proportions. The presence of 2-phenylethanol in Assyrtiko wines of Santorini was also noted by Symeou, Galiotou-Panayotou, Kechagia, and Kotseridis [20], Kallithraka et al. [32], and Tzamourani et al. [33]. However, Kechagia found isoamyl alcohol to be the main component of Assyrtiko wines, while they also detected the presence of 2-methyl-1-propanol [33]. Although the contribution of fusel alcohols to wine aroma is not obvious, they are well-known wine aroma components [34]. 2-methyl-1-propanol and 2-/3-methyl-1-butanol smell like nail varnish, whilst 2-phenylethanol has a rose-like scent. Finally, 1-hexanol is not a product of alcoholic fermentation but originates from the grape, smelling of freshly cut grass or green leaves and green parts of the plant, resulting in wines with a grassy smell and taste [34]. Its precepted threshold level is quite high, but the concentrations observed in the three wines were very low (less than 10 μg/L) [35]. Esters contribute significantly to the overall aromatic and flavor profile of wines by enhancing their complexity. Isoamyl, phenylethyl, and isobutyl acetates were identified in the Assyrtiko wines, which is in corroboration with Kechagia, Paraskevopoulos, Symeou, Galiotou-Panayotou, and Kotseridis [27], who also noted the presence of all three acetates, with isoamyl and phenylethyl acetates being the most important considering their concentration. The isoamyl acetate produced during alcoholic fermentation is an ester that gives banana and candy hints. Despite the low perception threshold, its concentration in the three wines is quite high, with Louro’s wine showing the highest level. Kallithraka, Christofi, Dimopoulou, Tsapou, and Papanikolaou [32] and Tzamourani, Paramithiotis, Favier, Coulon, Moine, Paraskevopoulos, and Dimopoulou [33] reported comparatively lower but at-the-same-level concentrations. Ethyl caprylate and ethyl caproate have tropical aromas and exhibited quite noticeable concentrations, especially in Ftelos wine. Finally, high concentrations of ethyl isobutyrate, having a perception threshold of 10 μg/L, were found, with the wine Papas having the highest amount of 163 μg/L, while Tzamourani, Paramithiotis, Favier, Coulon, Moine, Paraskevopoulos, and Dimopoulou [33] reported a concentration around 100 μg/L.

Terpenic compounds are characteristic of the primary aroma of certain varieties with an intense aroma. The grape-derived monoterpenoids of great importance for the aroma of wines are linalool and geraniol, giving floral and citrus (citrus) character. Monoterpenoids have a low perception threshold, but only linalool exceeds the limit, with Louros exhibiting the highest level. According to other research [36,37], monoterpenes—particularly linalool—benefitted from more growing degree days in Riesling and other varieties, as in our case, since grapes from the Louros site were collected later than others, continuing its ripening during a period with high air temperature. In conclusion, although we identified the same aromatic compounds in the three wines, the overall flavor perception of Assyrtiko wines is caused by their combination.

3.4. Sensory Evaluation Results

Wine tasters evaluated 13 different olfactory parameters of the three wine samples. Louros wine is slightly yellower compared to the other two wines, but the deviation is not large enough to be investigated further. According to Figure 2, the wine acidity is more noticeable in Louros and Papas, with Louros having the lowest pH. However, Ftelos wine has the highest sugar concentration, presenting a high deviation in this parameter, which probably affects the perception of acidity. Although Papas wine has fewer residual sugars, it achieved the highest mouthfeel perception score. Papas wine with the lowest alcohol degree perception comes from the vineyard with the highest altitude.

All wines were characterized by a high floral character according to GC-MS/MS, having a large amount of 2-phenylethanol, standing out with the Louros wine [27] (Figure 2). All three wines have increased citrus fruits, with the difference that Louros’s wine has the highest concentration, which is also consistent with the sensory results and the bibliography [27,32,33]. In addition, the wines produced in high-altitude sites are generally fresh, with high aromatic quality and a lower alcohol degree [28,38]. The aromatic profile of all three wines has a greater expression in citrus fruits (particularly in Louros wine) and floral aromas (particularly in Papas wine) (Figure 2). In second place are the white flesh and tropical fruity aromas (greater expression in Louros and Ftelos wines). Overall, Louros’ wine has greater complexity, harmony balance, and aftertaste than the other wines, which exhibited almost identical profiles. This fact may be due to the higher alcohol content, which gives a sense of volume to the wine.

4. Conclusions

Terroir-based winemaking from three different vineyards of the unique volcanic soils and microclimate of Santorini Island produced three distinct monovarietal wines. The vineyard at the lowest altitude, having soil with the lowest phosphorous level, gave wines with less fruit aromatic perception, as estimated by the judges. Grapes of the vineyard from the intermediate altitude were harvested 10 days earlier than those from the highest altitude. They gave wines with higher concentrations of volatile compounds with floral and citrus aroma perception. The sensory evaluation and the bibliography both corroborated this result. Additionally, given that linalool is considered a central monoterpene metabolite in grapes, based on our results, we could assume that it is associated with Assyrtiko grapes, and this could be an interesting piece of information for winemakers who want to extract the maximum aroma from the grapes and therefore could use the pre-fermentation extraction technique to produce Assyrtiko wine with more intense aromas of lemon, citrus, and bergamot.

Though fermentation and wine-handling processes were the same for the grapes of the three different vineyards, it is visible that altitude and soil composition may impact the technological maturity and, in turn, the chemical and sensory composition of wines. The vineyard with the latest ripening, probably because of the lower impact of the high temperature, gave wines with a good balance of acidity/alcohol and a more appreciated taste in terms of complexity, harmony, and aftertaste.

Assyrtiko grapes from the vineyard located at higher altitudes were harvested earlier, while grapes from the lowest altitude reached technological maturity later. Those grown at intermediate altitudes reached technological maturity much later. With the continuous rise in the temperature, estimating the appropriate timing of harvesting based only on sugar accumulation is not completely appropriate. Other parameters of berry composition, e.g., organic acids, aroma precursors, tannins, and color, as well as the volatile, chemical, and sensory profile of wines, should be considered.