Synergistic Strategy Against the Effects of Climate Change Using Non-Positioned Vegetation Training Systems and the Application of Kaolin in a Vineyard in a Semi-Arid Climate: Agronomic and Oenological Effects

Abstract

Climate change poses a major challenge for Mediterranean viticulture by accelerating ripening and reducing grape yield and quality. This study evaluated the synergistic effect of two adaptation strategies—non-positioned vegetation training (Sprawl) and foliar kaolin application—on the agronomic and oenological performance of Syrah cv. under semi-arid conditions over two consecutive seasons. Agronomic traits, bunch microclimate, and volatile composition of wines were determined. The combination of Sprawl and kaolin reduced bunch temperature by up to 2 °C, improved vine balance, and maintained optimal acidity and colour intensity. Wines from this treatment exhibited higher concentrations of esters and terpenes, generating more pronounced fruity, floral, and citrus aromas. Multivariate analysis of aroma series revealed clear differences between treatments and vintages, with 2025 showing stronger aromatic distinctions. Heatmap clustering confirmed that vintage was the main differentiating factor, followed by training system. These findings highlight the potential of integrating simple canopy management with reflective particle films to improve grape and wine quality under future Mediterranean conditions.

1. Introduction

The wine sector in Europe is of great socioeconomic importance. France, Spain, and Italy—together the main producers—account for 2.4 of the 7.1 million hectares of vineyards worldwide, and 111.6 of the 225.6 million hectolitres of global wine production including the most prestigious wines and wine-producing regions [1,2].

Climate change is profoundly altering viticulture in the Mediterranean basin and other parts of the world, where the main producing countries are located. This is mainly due to rising temperatures and reduced rainfall, which increase water stress and affect the accumulation of compounds responsible for yield and quality (sugar accumulation aroma, colour, and acidity of the grape) [3,4,5,6,7]. These changes bring forward the phenological cycle of the vine, causing an alteration in the date and duration of the different phenological stages. This alteration results in earlier ripening, with lower quality musts and wines due to the warmer climatic conditions in which they are produced. This lower quality is observed in musts and wines with lower acidity, higher pH, and a mismatch between technological, phenolic, and aromatic ripeness. As a result, prolonging ripening to compensate for both types of ripeness produces wines with higher alcohol content [7,8,9].

It is estimated that between 20% and 90% of wine-growing regions in Mediterranean countries (Spain, Greece and Italy) could be at risk of disappearing by the end of the century [7,10]. To mitigate these effects, strategies such as using adapted clones and rootstocks, reducing leaf mass, using alternative training systems, net-using or applying sunscreens such as kaolin have been proposed [7,11,12]. In recent years, even the use of agrivoltaics seems to be useful to address high temperatures and reduced water availability for grape and other fruit species [13].

Sprawl systems involve leaving the vegetation free in the training system, similar to bush vines, but with a training system that allows for complete mechanisation [14]. Sprawl systems have the advantage of protecting the bunches better from excessive temperature and solar radiation by shading them [15]. Furthermore, as the vegetation is not arranged in a parallelepiped as in vertical shoot positioning (VSP), the leaves receive direct sunlight and shade from adjacent leaves depending on the sun’s position at different times of day [16]. This has the advantage of increasing the photosynthetic efficiency of the leaves and reducing the degradation of anthocyanins in the grapes resulting from high temperatures and can even give wines a different wine signature [14,17].

Conversely, kaolin is commonly used on different types of crop, such as apples, pears, grapes, citrus and stone fruits. According to various authors, its use has numerous benefits for grapes. These include improved vine physiology, increased photosynthetic efficiency, reduced temperature and reduced damage from overexposure of the bunches [18,19,20,21,22,23]. The effects on wine production focus on general parameters, notably increases in acidity, ethanol and anthocyanins, and reductions in pH [24,25,26].

This study aims to assess the combined influence of Sprawl training and kaolin application on grapevine cv. Syrah physiology and aroma potential under semi-arid conditions in southern Spain.

2. Materials and Methods

2.1. Treatments Description

The treatment was carried out in a cv. Syrah commercial vineyard owned by Viñas de Alange, S.A. (Palacio Quemado Wineries) in Alange, Badajoz, Spain (38°40′15″ N 6°16′22″ W).

The vineyard employs a deficit irrigation regime with an allocation of around 700 m³/ha and has a density of 2778 vines/ha (1.2 row spacing × 3.2 m between rows). The vines are trained using the double Royat cordon system with six two-bud spurs per vine.

The study involved combining an asymmetrical, non-positioned vegetation training system (Sprawl) with the application of kaolin as sunscreen. Asymmetrical sprawl was implemented by allowing free canopy growth on the south side of the trellis at the start of the growing cycle and trimming occasionally to maintain the structure. The control training system was VSP (vertical shoot positioning).

Vegetation training systems (Figure 1) were complemented by a single application of kaolin (Sombreador Luqsa, LUQSA, Lleida, Spain) as sunscreen in veraison. This was done by spraying 1000 L per hectare of a 5% solution of the commercial product with 100% micronized kaolin. Combining both strategies would result in four treatments: (1) control (VSP); (2) sprawl; (3) VSP with kaolin application at veraison; and (4) sprawl with kaolin application at veraison.

To minimize the impact of soil and other uncontrollable factors, three replicates of 15 vines were established for each treatment and distributed across two rows.

The experimental treatments took place during the 2024 and 2025 vintages, encompassing the phenological stages from veraison (beginning of July) through harvest (August).

2.2. Climate Data Measurements

The climate study was conducted using data from a public weather station belonging to the Extremadura Irrigator Advisory Network (REDAREX) [27], which is in Villafranca de los Barros, 38°34′32.2″ N 6°20′54.8″ W, located 12.2 km from the test plot, both at a similar altitude (406 and 362 m, respectively). This analysis used the average temperature and precipitation values for the period 2011–2024, as well as specific data for the 2023/2024 and 2024/2025 seasons, which were analyzed separately.

2.3. Agronomic Measurements

A series of agronomic parameters were determined to ascertain how the vine developed under each condition. These parameters included vegetative, productive and microclimate data of the bunches.

For the vegetative parameters, the total number of shoots per vine and the exposed surface leaf area (SA) were measured. The exposed leaf area was determined by considering the VSP treatment as a parallelepiped and the Sprawl treatment as a quarter cylinder, where the exposed leaf area is the sum of the sides exposed to sunlight [28]. For the productive parameters, the number of bunches and yield per vine were determined. The average bunch weight was calculated by dividing the total harvest weight by the number of bunches per vine. The fertility index was also calculated as the ratio of the number of bunches and total shoots per vine. The SA-Yield ratio was also calculated by dividing leaf area by yield. This provides information on the balance between vegetative development and vine production.

Additionally, two sensors (Clima PKDL A1 datalogger, OWIM GmbH & Co., Neckarsulm, Germany) were placed in each treatment to record the temperature and relative humidity in the bunch area of the model vines (60 cm above the ground, shaded by leaves, Figure 2), to determine the effect of the different strategies on the bunches’ microclimate. The sensors recorded temperature and relative humidity every five minutes between veraison and harvest. The complete data can be found in Figures S1 and S2 in the Supplementary Materials.

2.4. Winemaking and Oenological General Parameters

All treatments were harvested on the same day as the winery’s commercial harvest (2 August 2024 and 4 August 2025). The grapes were collected in 15 kg boxes, after which they were destemmed and crushed, without the addition of SO₂.

Alcoholic fermentation was carried out using Velluto Evolution^® yeast (Lallemand Inc., 151 Skyway Avenue, Toronto, ON, Canada). Twenty-four hours after inoculation of the yeast for alcoholic fermentation, Lactivineibacillus vinearum (ML Prime^®, Lallemand Inc., 151 Skyway Avenue, Toronto, ON, Canada) was inoculated to carry out malolactic fermentation simultaneously. All fermentations were carried out in triplicate in stainless steel tanks thermostated at 21 ± 1 °C, one per agronomic plot, with daily punching down to promote the extraction of compounds from the skins. Fermentation was monitored by measuring density daily up to 995 g/L, and malolactic fermentation was monitored by measuring the malic acid content using the Reflectoquant™ system (Merck^®, Darmstadt, Germany) until the malic acid content fell below 0.25 g/L.

Once fermentation was complete, pressing was carried out using a hydraulic press at a maximum pressure of 2 bar to prevent breakage and extraction of compounds from the seeds. Then the wine was sulphited with 50 mg/L of SO₂ using potassium metabisulfite. The wine was then clarified and stabilised using cold stabilisation and the addition of 30 g/hL of vegetable protein (Proveget 100^®, Agrovin S.A., Alcázar de San Juan, Ciudad Real, Spain) and 20 g/hL of sodium bentonite (Maxibent G^®, Agrovin S.A., Alcázar de San Juan, Ciudad Real, Spain). The wine was then left to clarify for seven days at 4 °C.

After wine clarification, general oenological parameters (pH, titratable and volatile acidity, ethanol content, total polyphenol index and colour intensity) were determined using official wine analysis methods [29].

2.5. Volatile Compounds Determination

2.5.1. Major Volatile Compounds

Determination of the main compounds, including polyols, was performed using an Agilent HP 6890 Series II gas chromatograph equipped with a 50 m long, 0.25 mm internal diameter, 0.4 µm film thickness CP-WAX 57 CB capillary column and a flame ionisation detector. This methodology followed that of Peinado et al. [30]. For the analysis, 0.5 µL of pretreated wine was injected. Prior to injection, tartaric acid was removed by precipitation with 0.2 g of calcium carbonate and centrifugation at 300 g. To 10 mL of sample, 1 mL of 4-methyl-2-pentanol (1024 mg/L) was added as an internal standard. The equipment conditions were as follows: The split ratio was 30:1; the temperature programme started at 50 °C for 15 min, increasing by 4 °C per minute to 190 °C, which was maintained for 35 min. The injector was set at 270 °C and the detector at 300 °C. Helium acted as the carrier gas at a flow rate of 0.7 mL per minute for the first 16 min, increasing progressively to 1.1 mL per minute for the subsequent 52 min. Identification and quantification were performed by injecting standards under the same conditions as the samples.

2.5.2. Minor Volatile Compounds

The extraction and determination of minor volatile compounds were carried out in two phases, following the method described by López de Lerma et al. [31].

In the first extraction phase, the wine sample was diluted 1:10 in a hydroethanolic solution containing 12% ethanol (v/v). The pH was adjusted to 3.5 using 2.6 g/L of tartaric acid and 2.2 g/L of potassium bitartrate. A PDMS-coated stirring bar (Twister^®, 0.5 mm thick and 10 mm long, Gerstel GmbH, Mulheim an der Ruhr, Germany) was placed in a glass vial. The vial was then sealed with a teflon cap and stirred at 1500 rpm and 25 °C for 100 min. After this, the Twister was dried with lint-free paper and transferred to a thermal desorption tube for GC–MS analysis.

In the second phase, the determination stage, the desorption tube was introduced into the Gerstel TDS 2 system, which was coupled to the GC–MS. The Twister was then heated to release the extracts into a CIS 4 PTV injector containing a Tenax adsorption tube. Desorption was performed at 35 °C, increasing at a rate of 120 °C/min to reach 280 °C, which was maintained for 10 min with a helium flow rate of 3 mL/min. During this time, the injector was kept at 25 °C and then increased to 280 °C in splitless mode at a rate of 12 °C/s over a period of 7 min. The GC used an Agilent 19091S capillary column (30 m × 0.25 mm i.d., 0.25 µm) with helium as the carrier gas at a flow rate of 1 mL/min. The oven programme was set to 50 °C for 2 min, increasing at a rate of 4 °C/min to 190 °C, where it was held for 10 min. The mass detector operated in scan mode (39–300 m/z) to identify compounds based on their retention times and by comparing them with the Wiley 7N library and pure compounds. Quantification was performed using calibration curves of standard solutions that were treated in the same way as the samples, with ions selected in Hewlett-Packard Chemstation.

2.5.3. Aromatic Series Calculation

Aromatic series are defined as a group of volatile compounds with similar olfactory descriptors. The total OAV value of a series is calculated by the sum of the OAV values of its constituent compounds. The odour activity values (OAVs) of the volatile compounds were obtained by dividing their concentrations by their corresponding OPTs (olfactory perception thresholds). Nine aromatic series were identified: fruity, chemical, green, citrus, creamy, floral, green fruit, waxy, and honey. Note that a given volatile compound may belong to one or more of these series, depending on its specific sensory characteristics (Table S1).

2.6. Statistical Analysis

A two-way analysis of variance (MANOVA) was performed to evaluate the presence of significant differences between the experimental treatments, after confirming the hypothesis of homoscedasticity with Levene’s test and normality with Shapiro–Wilk’s test. The two vintages were analysed separately so as not to count the year as a factor of variation, taking the application of kaolin and the vegetation training system as variables. IBM SPSS Statistics 25 software (Armonk, NY, USA) was used for the statistical analysis.

Additionally, a cluster heatmap with correlation coefficients was created using the open-source Python 3.9.7 programming language in the Anaconda Jupyter Project environment (Anaconda Inc., Austin, TX, USA) to classify the different wines obtained. A heatmap with correlation coefficients was also created to analyse the relationship between agronomic and oenological parameters.

3. Results and Discussion

3.1. Climate Data

The climate data for the study site reveals significant summer aridity, which is a notable feature of the Mediterranean climate in which the vineyard is located. The average maximum and minimum temperatures are also noteworthy, with frequent heatwaves exceeding 40 °C [32].

During the study period (2023/24 and 2024/25 agricultural season), there was a significant increase in precipitation compared to the average, particularly in spring 2025 (Figure 3). In spring 2024 and 2025 (March to June inclusive), the total rainfall was 132 mm and 211 mm, respectively. This represents an increase of 5% and 67%, respectively, above the average of 126 mm for the same season in 2011–2023 period. This resulted in greater water availability for the vines. Similarly, summer temperatures increased significantly, particularly in the two months prior to harvest in both years, when numerous long-lasting heatwaves occurred, with temperatures exceeding 40 °C. Specifically, there were 19 and 36 days with maximum temperatures above 35 °C between June and July 2024 and 2025, respectively.

3.2. Agronomic Parameters

The agronomic parameters (

Table 1. Agronomic parameters and two-way analysis of variance (MANOVA) obtained for the different treatments.

			Shoots/Vine	Bunches/Vine	Yield	Bunches Weight	Fertility	Surface Area	Surface Area/Yield
					kg/Vine	g	Bunches/Shoot	m2/Vine	m2/kg
2024 Vintage	Sprawl		12 ± 2	16 ± 4	1.4 ± 0.5	92 ± 22	1.3 ± 0.4	2.5 ± 0.2	2.0 ± 0.9
	Sprawl + Kaolin		11 ± 2	17 ± 3	1.9 ± 0.6	108 ± 30	1.5 ± 0.3	2.3 ± 0.2	1.4 ± 0.5
	VSP		9 ± 1	15 ± 3	1.4 ± 0.3	93 ± 12	1.6 ± 0.4	2.7 ± 0.3	2.2 ± 0.7
	VSP + Kaolin		11 ± 2	16 ± 4	1.5 ± 0.5	88 ± 16	1.5 ± 0.3	2.8 ± 0.3	2.1 ± 0.6
	MANOVA	Training System	***	ns	ns	ns	ns	*	*
		Kaolin	ns	ns	ns	ns	ns	ns	ns
		TS × Kaolin	**	ns	ns	ns	ns	ns	ns
2025 Vintage	Sprawl		14 ± 1	21 ± 5	2.7 ± 0.6	132 ± 41	1.5 ± 0.3	3.2 ± 0.2	0.8 ± 0.2
	Sprawl + Kaolin		15 ± 2	25 ± 7	3.3 ± 1.3	126 ± 31	1.7 ± 0.3	3.4 ± 0.4	0.8 ± 0.2
	VSP		16 ± 3	27 ± 6	3.1 ± 0.9	116 ± 27	1.7 ± 0.3	3.2 ± 0.2	0.7 ± 0.2
	VSP + Kaolin		16 ± 2	25 ± 5	2.7 ± 0.7	111 ± 22	1.6 ± 0.3	3.3 ± 0.3	0.8 ± 0.2
	MANOVA	Training System	*	ns	ns	ns	ns	ns	ns
		Kaolin	ns	ns	ns	ns	ns	*	ns
		TS × Kaolin	ns	ns	ns	ns	ns	ns	ns

TS: Training System; VSP: Vertical shoot positioning; * p < 0.05; ** p < 0.01; *** p < 0.001; ns: not significant.

) show that the number of total shoots and the number of bunches remained stable between the different treatments in each season. However, there was a significant increase in the 2025 season compared to 2024. This was probably due to the improvement in the vine’s water status in 2024 compared to 2023 (a much drier year, 50.2 mm in spring), which favoured the fertility of the buds in 2025 during the differentiation occurring in the spring–summer 2024 [33,34], and, therefore, the number of bunches. In addition, higher rainfall in the spring of 2025 may have led to an improvement of bud sprouting and shoot development with also a higher cluster weight, this coincides with studies by other authors [33,34].

In the case of yield, although it is not statistically significant due to the variability between vines, there is a certain trend towards higher yields when both strategies are combined. This coincides with other studies such as Dinis et al. [21], who found increases in yield and improvements in the physiology of the Touriga Nacional grapevine variety in the Douro DOC in Portugal.

In terms of leaf area generated, the main differences were observed between seasons, rather than between the different treatments. This was due to greater water availability in the 2025 season, which was caused by high spring rainfall (see Figure 2). This greater water availability probably led to increased shoot growth [35,36,37].

Although more leaf area was generated in the 2025 season, the yield harvested was proportionally higher, resulting in a balance between leaf area and vine yield of around 0.8 m²/kg in all treatments. In the 2024 season, this ratio was between two and three times higher. This high vegetative-productive balance in favour of vegetative development in the 2024 season can result in significant increases in the sugar content of the grapes and consequently an earlier harvest [37]. Values around 1 m²/kg are optimal for proper ripening [33,35,38]. Values for the 2024 season are generally well above this optimum, but not for the 2025 season.

3.3. Bunches Microclimate

The bunches microclimate demonstrates the impact of the implemented strategies on temperature. The application of kaolin and the Sprawl system have both led to an average midday temperature drop of between 1 and 2 °C in the bunches area because of sunlight reflection and bunch shading (Figure 4). The decrease is greater in 2024 than in 2025, possibly due to the vine’s better water status in 2025, enabling greater transpiration, which is the vine’s main cooling mechanism [13].

Lower temperatures may favor metabolic processes such as the synthesis of anthocyanins and aromatic compounds, preventing their degradation due to high temperatures resulting from heatwaves in both seasons [38,39]. Lower temperatures can also reduce the degradation of acids during the ripening process, favoring wines with better pH and acidity [13,40]. In addition to their effects on grape metabolism, lower temperatures in the bunches area reduce water transpiration from the berries. Together with higher ambient humidity (Figure 5), this makes the berries less susceptible to raisining or heat stress, both of which negatively affect wine quality [39,41].

3.4. Oenological Parameters of the Obtained Wines

Looking at the oenological parameters shown in

Table 2. Oenological parameters and two-way analysis of variance (MANOVA) obtained for the analysed wines.

			Sugars ¥	pH	Titratable Acidity	Ethanol	Volatile Acidity	Anthocyanins	Colour Index	TPI
			g/L		g/L TH2	% v/v	g/L AcH	mg/L
2024 Vintage	Sprawl		244 ± 3	3.44 ± 0.01	6.86 ± 0.09	14.2 ± 0.1	0.33 ± 0.06	468 ± 15	28.2 ± 0.8	32.8 ± 0.6
	Sprawl + Kaolin		240 ± 4	3.37 ± 0.03	6.4 ± 0.2	14.1 ± 0.2	0.42 ± 0.03	434 ± 68	27.1 ± 0.6	32.0 ± 1.0
	VSP		254 ± 2	3.51 ± 0.03	5.79 ± 0.08	14.8 ± 0.1	0.44 ± 0.04	521 ± 24	33.0 ± 1.0	33.5 ± 0.5
	VSP + Kaolin		258 ± 6	3.44 ± 0.02	6.25 ± 0.04	15.0 ± 0.1	0.45 ± 0.05	478 ± 32	32.0 ± 1.1	33.4 ± 0.4
	MANOVA	Training System	***	**	***	***	*	*	***	*
		Kaolin	ns	**	ns	ns	ns	ns	ns	ns
		TS × Kaolin	ns	ns	***	ns	ns	ns	ns	ns
2025 Vintage	Sprawl		233 ± 4	3.43 ± 0.04	7 ± 0.1	13.6 ± 0.1	0.36 ± 0.03	537 ± 40	46.5 ± 0.6	30.8 ± 0.4
	Sprawl + Kaolin		235 ± 5	3.38 ± 0.03	7.46 ± 0.06	13.5 ± 0.1	0.32 ± 0.02	564 ± 39	43.9 ± 0.3	33.1 ± 0.2
	VSP		220 ± 5	3.19 ± 0.04	7.8 ± 0.3	12.6 ± 0.2	0.32 ±0.02	404 ± 22	40.2 ± 0.3	31.7 ± 0.2
	VSP + Kaolin		230 ± 3	3.21 ± 0.03	7.6 ± 0.3	13 ± 0.1	0.36 ± 0.03	465 ± 38	46.1 ± 0.6	34.1 ± 0.4
	MANOVA	Training System	ns	**	***	ns	***	***	***	**
		Kaolin	ns	ns	*	ns	*	***	***	**
		TS × Kaolin	ns	*	***	ns	ns	***	***	***

TS: Training System; VSP: Vertical shoot positioning; Sugars ¥: Initial sugar content of the must; TH₂: Tartaric acid; AcH: Acetic acid; TPI; Total polyphenol index; * p < 0.05; ** p < 0.01; *** p < 0.001; ns: not significant.

, it is worth noting that the VSP and VSP + Kaolin samples suffered slight raisining of the berries in the 2024 treatments as a result of the heatwave, due to the grapes’ greater exposure to and sensitivity to these conditions, in line with the observations of other authors [39,41]. Although ripening proceeded normally in 2025, VSP vines exhibited incomplete maturity compared with Sprawl, which may have caused incomplete seed ripening and harsh, green tannins. This could be due to greater chlorophyll degradation resulting from higher leaf temperatures caused by overexposure to sunlight [25].

A slightly lower pH (approximately 0.1 units) was observed in vines treated with kaolin. However, pH and acidity depend on factors beyond temperature, such as grape ripeness [42]. In 2024, the VSP and VSP + Kaolin treatments showed higher pH and lower acidity, likely due to slight berry raisining. In contrast, the 2025 results showed the opposite trend, with lower pH and higher acidity, consistent with the lower maturity of the grapes—reflected in their lower sugar content—compared with the Sprawl and Sprawl + Kaolin treatments [43].

A similar pattern was observed for ethanol: under greater water stress conditions in 2024, the Sprawl system showed greater resistance to berry raisining, resulting in lower ethanol levels compared with the control. This coincides with the findings of other studies, such as those of Winefield et al. [39]. Under conditions of water stress and high temperatures, the ethanol concentration in wine may exceed normal levels at normal maturity due to sugar concentration in the berry when it loses water through transpiration [41,44]. In contrast, in 2025, ripening was normal, and the Sprawl treatment enabled better grape ripening due to greater photosynthetic efficiency than the VSP treatments. This greater efficiency is because, in the Sprawl system, all the leaves receive solar radiation at different times of the day. In contrast, in the VSP treatment, the more compact layers of leaves favour overexposure of the outer leaves while the inner layers receive hardly any direct solar radiation [16].

A trend similar to that observed for ethanol was also evident for total anthocyanin content, colour intensity, and total polyphenol index. In the 2024 vintage, concentration effects predominated, likely resulting from slight berry desiccation in the VSP treatments, which led to wines with higher levels of these phenolic compounds. Conversely, in 2025, when grape ripening progressed under more typical conditions, higher values were recorded in the Sprawl and Sprawl + Kaolin treatments. This outcome can be attributed to reduced bunch exposure and, consequently, lower anthocyanin degradation—associated with limited sunlight exposure in the Sprawl system and lower cluster temperatures in the kaolin-treated vines [39,41]. This is consistent with a study by Frioni et al. [18] on Pinot Noir in central Italy, which found increased anthocyanin content in treatments with kaolin application. These findings may be extrapolated to other early-ripening varieties and potentially to some late-ripening ones, although, to the authors’ knowledge, no studies have yet examined these cases.

3.5. Volatile Wine Composition

3.5.1. Chemical Families

The volatile compounds analysed have been grouped into chemical families (

Table 3. Volatile compound concentrations (µg/L) and two-way analysis of variance (MANOVA) obtained after wine analysis.

	2024 Vintage							2025 Vintage
					MANOVA							MANOVA
	Sprawl	Sprawl + K	VSP	VSP + K	TS	K	TS × K	Sprawl	Sprawl + K	VSP	VSP + K	TS	K	TS × K
Major alcohols.	891 ± 8	889 ± 4	740 ± 7	712 ± 8	**	*	**	651 ± 2	708 ± 3	579 ± 7	534 ± 2	**	ns	**
Methanol	259 ± 16	359 ± 13	235 ± 19	242 ± 3	***	***	***	96 ± 4	120 ± 9	108 ± 11	93 ± 3	ns	ns	**
Propanol	66.7 ± 0.6	81 ± 3	76 ± 5	77 ± 10	ns	*	ns	67 ± 3	64 ± 2	79 ± 2	88 ± 4	***	ns	**
Isobutanol	67 ± 3	44.4 ± 0.9	45 ± 4	46 ± 4	***	***	***	45 ± 5	51 ± 2	43 ± 3	40 ± 1	**	ns	*
2-methylbutanol	66 ± 6	46 ± 5	56 ± 7	44 ± 6	ns	**	ns	59 ± 2	64 ± 2	41 ± 2	30 ± 6	***	ns	*
3-methylbutanol	373 ± 23	277 ± 7	263 ± 21	244 ± 25	***	**	*	309 ± 8	323 ± 7	265 ± 22	246 ± 6	***	ns	ns
2-Phenylethanol	58 ± 9	80 ± 8	65 ± 5	58 ± 5	ns	ns	**	74 ± 2	86 ± 2	44 ± 3	37 ± 1	***	*	***
Minor alcohols.	2576 ± 18	1580 ± 45	2355 ± 80	1984 ± 141	ns	***	***	2196 ± 131	2391 ± 113	2357 ± 164	2012 ± 54	ns	ns	**
Hexanol	2319 ± 16	1326 ± 39	2007 ± 75	1774 ± 149	ns	***	***	2079 ± 141	2289 ± 109	2250 ± 171	1910 ± 51	ns	ns	**
2-ethyl-1-hexanol	39 ± 2	47 ± 1	53 ± 5	51 ± 5	**	ns	ns	21.7 ± 0.5	24.4 ± 0.9	22 ± 2	21 ± 2	ns	ns	ns
Octanol	215 ± 20	203 ± 11	207 ± 23	154 ± 4	*	**	ns	94 ± 11	75 ± 6	83 ± 9	79 ± 6	ns	*-	ns
Dodecanol	0.8 ± 0.06	2.8 ± 0.3	86 ± 6	2.9 ± 0.2	***	***	***	1.13 ± 0.03	1.7 ± 0.1	1.5 ± 0.2	1.1 ± 0.2	ns	ns	***
Farnesol	1.6 ± 0.1	0.18 ± 0.01	1.35 ± 0.08	2 ± 0.2	***	***	***	0.29 ± 0.04	1.1 ± 0.2	0.14 ± 0.02	0.12 ± 0.02	***	***	***
Major esters	175.5 ± 0.7	170.1 ± 0.4	171 ± 3	174 ± 1	ns	ns	ns	148.3 ± 0.8	136.2 ± 0.6	144 ± 1	155 ± 1	*	ns	*
Ethyl acetate	63 ± 2	61 ± 3	78 ± 2	76 ± 5	***	ns	ns	78 ± 3	74 ± 3	78 ± 3	83 ± 2	*	ns	*
Ethyl lactate	48 ± 2	32 ± 2	35.8 ± 0.9	33 ± 7	*	**	*	24.4 ± 0.9	25 ± 1	28.8 ± 0.8	30 ± 2	***	ns	ns
Diethyl succinate	65 ± 4	78 ± 2	57 ± 8	66 ± 5	*	**	ns	46 ± 1	38 ± 2	37 ± 1	42 ± 4	ns	ns	**
Minor esters:	3966 ± 140	2571 ± 28	3267 ± 47	3327 ± 80	ns	***	***	10,246 ± 410	9559 ± 473	6367 ± 241	6971 ± 131	***	ns	ns
Ethyl propanoate	nd	nd	nd	nd				62 ± 6	59 ± 3	45 ± 1	43 ± 2	***	ns	ns
Ethyl isobutanoate	26 ± 1	26 ± 1	31.1 ± 0.1	22.7 ± 0.9	ns	***	***	nd	nd	nd	nd
Ethyl butanoate	198 ± 10	140 ± 4	153 ± 2	161 ± 7	*	***	***	237 ± 21	207 ± 10	157 ± 4	180 ± 4	***	ns	**
Butyl acetate	0.45 ± 0.05	0.57 ± 0.02	0.73 ± 0.08	0.7 ± 0.06	***	ns	ns	2.2 ± 0.2	1.5 ± 0.2	1.6 ± 0.1	2.65 ± 0.05	*	ns	***
Ethyl 2-methylbutanoate	1.2 ± 0.07	1.7 ± 0.2	2.1 ± 0.2	1.6 ± 0.2	**	ns	**	2.3 ± 0.2	2.2 ± 0.3	1.2 ± 0.1	1.76 ± 0.06	***	ns	**
Ethyl 3-methylbutanoate	3.7 ± 0.2	4.1 ± 0.1	5.1 ± 0.3	4.4 ± 0.3	***	ns	**	2.1 ± 0.2	2.72 ± 0.03	2.1 ± 0.2	2 ± 0.2	**	*	*
Isoamyl acetate	2408 ± 74	1254 ± 31	1982 ± 15	1652 ± 38	ns	***	***	5834 ± 122	5069 ± 263	3808 ± 121	4580 ± 132	***	ns	***
Ethyl hexanoate	284 ± 13	326 ± 2	310 ± 10	392 ± 27	**	***	ns	661 ± 36	562 ± 12	577 ± 16	553 ± 13	**	**	*
Hexyl acetate	28.2 ± 0.6	25.1 ± 0.8	18.3 ± 0.6	27 ± 2	***	**	***	253 ± 9	230 ± 4	272 ± 8	252 ± 6	**	**	ns
Ethyl heptanoate	0.13 ± 0.01	0.19 ± 0.02	0.15 ± 0.01	0.48 ± 0.03	***	***	***	0.23 ± 0.01	0.33 ± 0.01	0.23 ± 0.03	0.22 ± 0.01	***	**	***
Ethyl octanoate	77 ± 2	175 ± 9	39 ± 2	236 ± 27	ns	***	***	67 ± 4	222 ± 4	98 ± 5	76 ± 2	***	***	***
Ethyl phenylacetate	6.4 ± 0.3	2.2 ± 0.1	1.6 ± 0.1	2.1 ± 0.2	***	***	***	317 ± 14	326 ± 19	168 ± 9	139 ± 3	***	ns	*
2-Phenylethyl acetate	729 ± 85	435 ± 10	570 ± 13	641 ± 37	ns	**	***	2549 ± 268	2602 ± 181	1049 ± 94	902 ± 37	***	ns	*
Ethyl decanoate	106 ± 8	93 ± 3	64 ± 2	108 ± 10	**	**	***	199 ± 5	213 ± 8	145 ± 5	183 ± 2	***	***	**
2-Phenylethyl butanoate	17 ± 1	5.6 ± 0.2	11.1 ± 0.6	12.8 ± 0.9	ns	***	***	25 ± 0.1	25 ± 1	9 ± 0.4	7 ± 0.4	***	ns	ns
Ethyl tetradecanoate	24 ± 2	27 ± 1	nd	20 ± 3	ns	ns	**	11.7 ± 0.3	12.3 ± 0.5	12 ± 0.5	17 ± 1	***	***	**
Phenethyl benzoate	2.67 ± 0.03	2.7 ± 0.1	3.4 ± 0.1	3.3 ± 0.1	***	ns	ns	1.4 ± 0.03	1.38 ± 0.04	1.4 ± 0.1	1.47 ± 0.05	ns	ns	ns
Ethyl hexadecanoate	54 ± 6	53 ± 4	56 ± 4	41 ± 7	ns	ns	ns	22.5 ± 0.5	23 ± 1	22 ± 1	30 ± 3	**	**	**
Major aldehydes:	139 ± 14	142 ± 11	138 ± 15	132 ± 5	ns	ns	ns	98 ± 3	98 ± 3	105 ± 6	108 ± 7	ns	ns	ns
Acetaldehyde	139 ± 14	142 ± 11	138 ± 15	132 ± 5	ns	ns	ns	98 ± 3	98 ± 3	105 ± 6	108 ± 7	ns	ns	ns
Minor aldehydes:	49 ± 3	36 ± 2	37 ± 1	45.7 ± 0.8	ns	ns	***	24 ± 1	28 ± 2	19.9 ± 0.4	24.1 ± 0.3	*	*	ns
Benzaldehyde	2 ± 0.2	0.79 ± 0.08	2.11 ± 0.08	1.8 ± 0.1	***	***	***	2.1 ± 0.3	2.4 ± 0.2	2.35 ± 0.09	1.7 ± 0.1	ns	ns	**
Heptanal	1.1 ± 0.1	1.13 ± 0.09	0.7 ± 0.1	0.82 ± 0.09	***	ns	ns	0.13 ± 0.01	0.15 ± 0.03	0.58 ± 0.03	0.22 ± 0.03	***	***	***
Octanal	1.46 ± 0.07	1.9 ± 0.2	1.8 ± 0.1	1.9 ± 0.1	ns	*	ns	4.0 ± 0.1	3.64 ± 0.08	4.3 ± 0.1	4.0 ± 0.1	**	**	ns
Nonanal	4.8 ± 0.6	6.3 ± 0.4	5.6 ± 0.2	5.5 ± 0.2	ns	*	*	1.9 ± 0.3	3.6 ± 0.4	2.7 ± 0.3	3.7 ± 0.3	*	***	ns
Decanal	7.3 ± 0.6	7.6 ± 0.9	7 ± 1	8.3 ± 0.9	ns	ns	ns	2.3 ± 0.2	4.1 ± 0.3	3 ± 0.3	5 ± 0.6	**	***	ns
Phenylacetaldehyde	32 ± 2	18.2 ± 0.8	19.9 ± 0.6	27 ± 2	ns	**	***	17 ± 2	17 ± 2	9.8 ± 0.9	12 ± 1	***	ns	ns
Major ketones.	68 ± 3	69 ± 3	42 ± 1	51 ± 5	***	*	ns	35 ± 1	29 ± 2	51 ± 2	59 ± 6	***	ns	**
Acetoin	68 ± 3	69 ± 3	42 ± 1	51 ± 5	***	*	ns	35 ± 1	29 ± 2	51 ± 2	59 ± 6	***	ns	**
Minor ketones:	3.4 ± 0.2	2.4 ± 0.2	3.08 ± 0.06	3.1 ± 0.3	ns	**	**	7 ± 0.4	8.3 ± 0.3	7.9 ± 0.4	6.8 ± 0.2	ns	ns	***
Benzophenone	0.86 ± 0.06	0.77 ± 0.07	0.69 ± 0.05	0.78 ± 0.07	ns	ns	*	2.3 ± 0.2	2.5 ± 0.1	2.1 ± 0.1	1.5 ± 0.2	***	*	**
3-heptanone	0.15 ± 0.02	0.07 ± 0.01	0.27 ± 0.05	0.24 ± 0.02	***	*	ns	2.9 ± 0.2	2.86 ± 0.07	2.9 ± 0.3	3 ± 0.3	ns	ns	ns
Acetophenone	2.4 ± 0.2	1.6 ± 0.1	2.12 ± 0.09	2.1 ± 0.3	ns	**	*	1.8 ± 0.2	2.9 ± 0.3	2.9 ± 0.3	2.3 ± 0.3	ns	ns	**
Lactones	10 ± 0.7	8.8 ± 0.6	10 ± 1	10.2 ± 0.7	ns	ns	ns	10.9 ± 0.8	10.1 ± 0.8	9.4 ± 0.8	4.5 ± 0.4	***	***	**
γ-nonalactone	8.7 ± 0.6	7.3 ± 0.5	9 ± 1	8.9 ± 0.8	ns	ns	ns	9.4 ± 0.7	8.9 ± 0.9	8.7 ± 0.8	3.4 ± 0.3	***	***	***
γ-decalactone	1.33 ± 0.09	1.5 ± 0.1	1.48 ± 0.09	1.3 ± 0.1	ns	ns	*	1.5 ± 0.1	1.1 ± 0.1	0.7 ± 0.1	1.1 ± 0.1	**	ns	**
Terpenes and norisoprenoids	51 ± 3	50 ± 2	54 ± 4	50.06 ± 0.09	ns	ns	ns	50 ± 2	55 ± 1	38 ± 4	33 ± 2	***	ns	**
Limonene	16.8 ± 0.8	17 ± 1	15 ± 1	16 ± 1	ns	ns	ns	17 ± 1	21 ± 1	16 ± 2	10 ± 1	***	ns	**
β-citronellol	25 ± 3	22 ± 2	30 ± 3	23 ± 2	ns	**	ns	23 ± 2	23.9 ± 0.6	12 ± 1	11.8 ± 0.9	***	ns	ns
β-damascenone	4.1 ± 0.2	4.6 ± 0.4	2.74 ± 0.05	3.6 ± 0.3	***	*	**	6.7 ± 0.2	7 ± 0.2	6.2 ± 0.4	7.2 ± 0.1	ns	**	ns
Z-geranyl acetone	2.3 ± 0.2	2.09 ± 0.09	2.1 ± 0.1	2.23 ± 0.03	ns	ns	ns	2.02 ± 0.03	2.2 ± 0.2	2.11 ± 0.08	2.1 ± 0.1	ns	ns	ns
E-methyldihydrojasmonate	3.2 ± 0.3	3.8 ± 0.2	4.1 ± 0.4	4.7 ± 0.5	**	*	ns	1.3 ± 0.1	0.94 ± 0.07	1.6 ± 0.2	1.4 ± 0.2	**	**	ns

TS: Training System; K: Kaolin application; VSP: Vertical shoot positioning * p < 0.05; ** p < 0.01; *** p < 0.001; ns: not significant.

). A decrease is observed in the Sprawl wines for the major higher alcohols compared to the VSP. These higher alcohols are produced by yeast as a product of nitrogen metabolism and are linked to certain precursor amino acids. The compounds that contribute most, 2- and 3-methylbutanol and isobutanol, are related to the amino acids isoleucine, leucine and valine, respectively. An increase in these amino acids can cause an increase in the corresponding higher alcohol [45]. Some authors have found decreases in these amino acids because of higher grape temperatures, despite an increase in total amino acid content [46]. This could explain the decrease observed in the most exposed grapes (VSP), which reached higher temperatures during ripening (Figure 3), although this depends on the grape variety [46]. Among the minor alcohols, notable differences are observed in hexanol, which is the main compound that makes up this group. This compound, which has a vegetal or herbaceous aroma, and it is formed from grape membrane lipids by lipoxygenase enzymes during pre-fermentation processes [47]. Its reduction is particularly notable in the 2024 vintage when kaolin was used, which may be linked to greater grape ripeness [48].

The esters highlight the influence of the treatments carried out, as well as the vintage. There was an increase of between 91% and 270% between the 2024 and 2025 vintages. The main compounds in this group are isoamyl acetate, 2-phenylethyl acetate, ethyl butanoate and hexanoate. The reduction in esters may be due to the grapes being slightly overripe in the 2024 vintage because of lower rainfall. This would be consistent with the findings of Trujillo et al. [49], who observed significant reductions in esters such as ethyl butanoate and hexanoate in Merlot wines produced from overripe grapes. There was also a reduction of close to 50% in isoamyl and 2-phenylethanol acetates.

Unlike the other compounds, terpenes and norisoprenoids come directly from the grape and are therefore more influenced by growing conditions [50,51]. They have fruity and floral aromas [47]. The presence of limonene and β-citronellol is noteworthy as these compounds also have citrus and floral aromas [47]. Terpene concentration is favoured in the sprawl treatments, which may be due to the lower temperatures recorded in the bunches area (Figure 3). Terpene synthesis is favoured by an increase in temperature up to the 40 °C threshold, above which extreme heat can negatively affect synthesis, as seen in the VSP treatments in this study [52]. β-damascenone shows significant differences between the two vintages, possibly due to excessive ripening in the 2024 vintage, particularly in the control treatment (VSP). Similar results were observed by Talaverano et al. [53], who found significant reductions in this compound as the grapes became riper, regardless of the amount of water stress the vines were subjected to.

3.5.2. Aromatic Series Values

The olfactory activity value (OAV) is used to evaluate the relative influence of each compound on the overall aroma. It is considered a very useful analytical parameter [54] and has been used by numerous authors [9,55,56]. Generally, volatile compounds with an OAV greater than one are considered to potentially define the aromatic character of wine. However, this depends not only on the presence of volatile compounds, but also on the interactions between them [57].

A total of sixteen volatile compounds were identified, including isoamyl acetate, ethyl hexanoate, ethyl octanoate, ethyl butanoate, 2-phenylethyl acetate, ethyl propanoate, phenylacetaldehyde, octanal, decanal, limonene, ethyl phenylacetate, ethyl decanoate, β-damascenone, nonanal, 2-phenylethanol, and 3-methyl-1-butanol. Most of these compounds have been reported in previous studies as key contributors to the characteristic aroma profile of aromatic wines [55]. Fruity, floral, citrus, and green sensory attributes are mainly associated with these compounds.

The volatile compounds were grouped into aromatic series according to their descriptors. The value of a series is obtained by summing the OAVs of its constituent compounds. Nine series were identified: fruity, green fruit, green, creamy, citrus, chemical, honey, waxy and floral (

Table 4. Aromatic series and two-way analysis of variance (MANOVA) obtained for the analysed wines.

			Fruity	Green Fruit	Green	Creamy	Citrus	Chemistry	Honey	Waxy	Floral
2024 Vintage	Sprawl		132 ± 1	22 ± 1	8.8 ± 0.5	1.1 ± 0.1	10.0 ± 0.7	32.9 ± 1	11.0 ± 0.5	22.1 ± 0.5	10.2 ± 1.1
	Sprawl + Kaolin		113 ± 2	25 ± 2	5.1 ± 0.2	1.0 ± 0.2	11.1 ± 0.8	30.3 ± 0.5	6.3 ± 0.2	42 ± 1	11.1 ± 0.8
	VSP		110 ± 2	24 ± 1	5.5 ± 0.1	0.8 ± 0.1	10 ± 1.3	29.8 ± 1.2	7.3 ± 0.1	14.3 ± 1.2	10.1 ± 0.5
	VSP + Kaolin		144 ± 6	30 ± 2	7.4 ± 0.4	0.9 ± 0.1	11.2 ± 1	28.6 ± 0.5	9.4 ± 0.4	55 ± 5	9.7 ± 0.4
	MANOVA	TS	ns	**	*	**	ns	**	ns	ns	ns
		Kaolin	**	***	**	ns	ns	**	***	***	ns
		TS × Kaolin	***	ns	***	*	ns	ns	***	***	ns
2025 Vintage	Sprawl		277 ± 9	49 ± 3	4.9 ± 0.4	0.7 ± 0	5.9 ± 0.4	28.5 ± 0.9	51 ± 2	16.4 ± 1	20.5 ± 1.2
	Sprawl + Kaolin		274 ± 10	42 ± 1	5 ± 0.5	0.7 ± 0	8.2 ± 0.1	29.5 ± 0.9	53 ± 3	48.9 ± 0.8	22 ± 0.9
	VSP		204 ± 6	42 ± 1	3.3 ± 0.2	0.8 ± 0	6.8 ± 0.2	26.5 ± 1.0	27 ± 1	22.7 ± 0.7	10.5 ± 0.4
	VSP + Kaolin		224 ± 5	41 ± 1	3.8 ± 0.3	0.7 ± 0.1	8.1 ± 0.5	25.2 ± 0.7	24 ± 1	20.2 ± 0.9	9.3 ± 0.2
	MANOVA	TS	***	**	***	**	**	***	***	***	***
		Kaolin	ns	**	ns	**	ns	ns	ns	***	ns
		TS × Kaolin	ns	*	ns	ns	ns	ns	ns	***	*

TS: Training System; VSP: Vertical shoot positioning; * p < 0.05; ** p < 0.01; *** p < 0.001; ns: not significant.

).

The fruity, green fruit and waxy series emphasise the differences in Sprawl compared to the control (VSP), and in the 2025 vintage compared to the 2024 one. They are mainly composed of esters. Notably, among these esters are isoamyl acetate and the ethyl esters of hexanoic, octanoic, and butanoic acids. As discussed in Section 2.5.1, certain authors have observed notable reductions in overripe grapes compared to those at normal ripeness, which could account for the variations between the two vintages.

The differences are smaller in the green and creamy series, but the Sprawl treatments stand out compared to VSP. In these series, phenylacetaldehyde and γ-nonalactone are notable, although the latter does not exceed the OAV unit.

In the citrus series, increases are observed with the application of kaolin, though these are only significant in the 2025 vintage. The main compounds in this series are limonene, decanal and nonanal aldehydes, and their increase may be due to lower temperatures in the vine and bunches area resulting from the application of sunscreen, which increases enzymatic berry activity at very high temperatures, as observed in the present study [52].

The major higher alcohols are the primary compounds in the chemistry series. The VSP treatment shows lower values than the Sprawl, with no obvious impact from the use of kaolin. This may be related to the presence of specific amino acids and how temperature affects them, as discussed in Section 2.5.1.

The floral aromatic series was strongly influenced by the vintage. While no major differences were observed in 2024, clear variations emerged in 2025, when grape ripening proceeded under normal conditions. In this latter vintage, the Sprawl treatments exhibited a significant increase in floral compounds compared with the VSP system, primarily due to higher concentrations of 2-phenylethanol and its ester, 2-phenylethyl acetate. The latter compound is formed through the esterification of 2-phenylethanol with acetic acid [40]. Yeast produces 2-phenylethanol primarily as a by-product of amino acid metabolism, as discussed in Section 2.5.1 for other higher alcohols. Moukarcel et al. [58] reported significant decreases in phenylalanine—the amino acid precursor of 2-phenylethanol—under elevated bunch temperatures. However, other studies, such as that of Wang et al. [48], have shown that grape variety exerts a strong influence on the response of this amino acid to temperature variations. Another factor that may contribute to higher concentrations of 2-phenylethanol and its ester, 2-phenylethyl acetate, is grape maturity. In the 2025 vintage, grapes from the Sprawl treatments achieved a higher degree of maturity than those from the VSP system, in agreement with observations reported by other authors [59].

3.6. Multivariate Analysis

A multivariate method was applied to create a star plot with nine axes, each corresponding to a distinct aroma series (see Figure 6). Before plotting, the data were standardized to ensure equal axis lengths, with unit values representing the mean of each series. This normalization enables direct comparison among the different aroma series. The resulting visualization allows for the identification of predominant aroma attributes within samples, highlights similarities across observations, and supports the delineation of potential clusters [60].

In 2024 (Figure 6a), the samples show relatively balanced aroma distributions, with Waxy and Green notes being slightly more prominent in the Sprawl and VSP + Kaolin treatments, respectively. The overall aromatic expression is moderate, and differences between treatments are subtle, suggesting that environmental or viticultural factors during this vintage exerted limited influence on the development of specific aroma series.

In 2025 (Figure 6b), clearer distinctions among treatments emerge. The Sprawl + Kaolin and VSP + Kaolin samples exhibit notably higher intensities in Fruity, Green fruit, and Floral attributes compared to 2024, indicating an enhancement of fresh and floral aromatic components. Conversely, Sprawl samples maintain stronger Waxy and Honey notes, suggesting the persistence of more mature or lipid-derived volatiles. The increased spread of the curves relative to the median highlights greater aromatic differentiation among treatments in 2025, possibly reflecting distinct climatic conditions or the cumulative effect of canopy management and kaolin application.

Overall, the multivariate aroma profiles suggest that kaolin treatment and trellis system interact to modulate specific aroma series, with their effects becoming more pronounced in 2025. The normalization of data across aroma axes facilitates the identification of dominant aromatic trends and potential clustering among treatments over time.

3.7. Cluster Heatmap

To determine the integrated effect of all the analysed parameters, a cluster analysis was performed using a heat map to improve explanation of the obtained results (Figure 7).

The heatmap was created to facilitate explanation of the cluster analysis results. To this end, each variable was normalised to avoid biases based on the units of each parameter. The value and colour of each variable are represented in the centre of the corresponding square: blue for negative values and red for positive values.

The analysis revealed a clear influence of vintage on wine composition, highlighting the critical role of climatic conditions in the vineyard. Variables such as titratable acidity, colour index, and the Fruity, Green Fruit, and Citrus aromatic series contributed most strongly to this differentiation, all showing markedly higher values in the 2025 vintage. A secondary level of distinction was observed among the groups defined by training system in 2025, primarily driven by anthocyanin content and the Honey, Waxy, and Floral aromatic series. In this context, the Sprawl system exhibited higher values than the VSP system.

In the group corresponding to the 2024 treatments, the secondary level of grouping was less distinct. However, it appeared to be more strongly influenced by the application of kaolin. The main differences within this group were associated with higher values of the Waxy aromatic series and a reduction in pH in the kaolin-treated samples.

4. Conclusions

Using non-positioned vegetation training (sprawl) alongside foliar kaolin application proved to be an effective adaptation strategy against climate change for Vitis vinifera cv. Syrah in a semi-arid environment. This synergy lowered bunch zone temperature by up to 2 °C, improved vine balance and maintained acidity levels during periods of intense heat. Compared to wines produced using the conventional vertical shoot positioning (VSP) system, wines from the Sprawl + Kaolin treatment showed higher color intensity, total polyphenols, and aromatic complexity, with a richer volatile profile containing more fruity, floral, and citrus notes due to higher levels of esters and terpenes.

Multivariate aroma analysis revealed distinct fingerprints for each treatment. In 2025, wines from the Sprawl + Kaolin treatment were dominated by fruity, green fruit, and floral aromas, while wines from the Sprawl treatment alone were linked to waxy and honey notes—a differentiation absent in 2024, indicating stronger effects under balanced cli-mate conditions.

Cluster heatmap analysis confirmed two main groupings by vintage and training system. Titratable acidity, colour intensity and fruity/citrus aromas separated the vintages, whereas anthocyanins and the honey, waxy and floral series distinguished the canopy systems.

Overall, the Sprawl + Kaolin approach enhances the resilience of vineyards and the quality of wine produced, mitigating the effects of heat stress and stabilising the development of aromas. This makes it a sustainable strategy for viticulture in Mediterranean and other warm climates.