Terroir Traceability in Grapes, Musts and Gewürztraminer Wines from the South Tyrol Wine Region

Abstract

This study arose from the need to relate specific terroir aspects with experienced sensory properties of Gewürztraminer wines from Tramin (northern Italy). A multidisciplinary approach was used to investigate seven vineyards’ ecological characteristics, including geology and geographical features. A geopedological method using Vineyard Geological Identity (VGI) and Solar Radiation Identity (SRI) for topoclimatic classification, as well as multi-parameter measuring stations of air and soils, elicited analytical data for qualitative and quantitative terroir characterisations. Furthermore, wide-ranging and targeted oenological and chemical analyses were conducted on grapes, musts and wines to correlate their biochemical compositions with the measured terroir conditions. The study identified strong connections between vineyard geo-identity and wine mineral fingerprint, confirming mineral traceability of Rb/Sr ratio and of some minerals common to the local geology, such as Ba, Rb, Mn and Be. In particular, the most differing geo-mineral and physical soil conditions of two studied vineyards are apparent in the oenological components, flavours and aromas of their musts and finished wines. Amino acids, primary varietal aromas and polyphenols, thiol compounds with tropical scents, phenolic compounds with spicy notes and terpenic compounds, such as geraniol and citronellol, were related differently between fine-textured, more siliceous soils of glacial origin and coarser-textured, more dolomitic soils of local debris flow origin.

1. Introduction

Terroir is known to influence the characteristics of wine in its many components [1,2]. In nature, wine styles are dictated by complex physiological processes during plant growth, which defines the grapes’ chemical composition. Several of these processes are closely linked to terroir and some environmental factors that characterise a particular vineyard. Soil and climate conditions play a pivotal role in creating a wine’s sensory attributes, quality and style [3,4,5]. The winegrower is also a factor in shaping a wine’s style and quality by choosing vine variety and vineyard management practices adapted to local soil and climate conditions [6]. Using appropriate winemaking techniques, winemakers not only translate berry composition into the best possible wine but also influence the so-called terroir effect, either positively or negatively [7]. However, interactions among environmental factors (climate and soil), grape variety, vine management and winemaking practices are quite complex, but they can be better assessed by breaking them down into measurable factors and sub-factors. This research uses a pluridisciplinary approach, which also includes the study of the geological origin of landforms and soils’ geopedology; it aims to interpret the differences in flavour typicality of the Gewürztraminer cultivar growing in a historically established geographical location and terroir in the South Tyrol wine region.

The Gewürztraminer is sensitive to abiotic stresses, with very typical biochemical characteristics and primary aromatic compounds [8]. This experimental study quantitatively evaluates abiotic sub-factors to examine the effects of soil and climate features on major aroma compounds expressed in wine to better understand whether and how terroir can shape Gewürztraminer wine typicity near the village of Tramin, Italy.

Previous and recent reports have focused on a site’s natural resources. Major parameters determining terroir expression can be quantified and even spatialised on a vineyard scale [9,10]. Climate components can be broken down into sub-factors that can affect wine’s sensory attributes: air temperature [11,12,13], solar radiation [7,14,15,16], rainfall and vine water status [17,18,19]. According to Choné et al. and Withe [20,21], soil temperature, which can be affected by the nature of the soil surface, as well as the physical and mineralogical features, land slope and aspect, are also important.

Whereas many authors have studied and acknowledged a specific effect of soil on terroir expression [9,22,23,24,25], the aspect of geological origin and geopedological soil parameterisation, which in complex mountainous environments underlies the complexity of the interactions determining soil’s terroir, has been little studied in the past [5,26].

In the Italian Alps, due to their multifaceted geographical variability, some factors and sub-factors of terroir, such as soil and soil geology, climate, geographical location and topoclimate, more commonly affect grape and wine characteristics [27]. For example, silicate minerals express acidic soils, dolomite minerals provide moderately alkaline pH values, and limestone favours alkaline soils. The content of clays and phyllosilicate minerals supports cation exchange capacity and soil moisture [28,29]. These are geopedological elements that fundamentally influence plants’ biosynthetic activity and fruit quality [30,31]. Therefore, exhaustive knowledge of a soil’s mineralogical component carries essential meaning in terroir investigation, particularly for a complex ecological environment such as the South Tyrol wine region.

The main objective of this study is to describe a complex terroir and test new methods to evaluate whether there is evidence of the organoleptic variability measured between wines over decades of local winemaking tradition. Of particular importance in the local terroir are the highly variable geopedological conditions and geographical landforms, which are studied using the Solar Radiation Identity (SRI) [26] and Vineyard Geological Identity (VGI) [5] methods. Whether other conditions (e.g., air and soil temperature, rainfall, human practices) of abiotic stress can be considered important in the studied vineyards is also evaluated.

2. Materials and Methods

The terroir features of a small and very typical production area of the Gewürztraminer grape variety were analysed. The studied area is located near the village of Tramin, in the Alto Adige-Südtirol DOC wine region, where the multiplicity of Alpine mountain landscapes and their geographical characteristics create extremely fertile conditions for biodiversity [32,33,34]. In our experiments, terroir factors–particularly soil geology, light, soil and air temperature, and soil moisture–were investigated separately. However, vineyard practices and winemaking techniques were kept strictly constant in all the vineyards. Therefore, they did not represent variable elements in this research. Furthermore, an experimental investigation was conducted on abiotic environmental elements, with a primary focus on qualitatively and quantitatively distinguishing the indicators of typical soil and microclimate vineyards in a very complex and varied geopedological environment: in this case, the Alpine region. Seven vineyards were analysed, documenting the possibility of tracing the soil-marking mineral elements and the possible effects of terroir factors on wine products. The vineyard classification techniques followed the VGI and SRI methods to facilitate a detailed qualitative–quantitative definition of the primary ecological factors of a geopedological and topoclimatic nature, and to obtain a meticulous mapping of soil functional properties at a reasonable cost and time, particularly where spatial variations in soils are more complex (e.g., near Tramin). The climate measurements in the vineyards were conducted using multi-parameter measuring stations, which continuously monitored soil and air.

To assess the possible correlations between terroir ecological indicators and vineyard products’ bio-fingerprints, oenological–chemical analyses were conducted on grapes, musts and wines using targeted procedures. The primary indicators of the more typical Gewürztraminer cultivar were examined, focusing mainly on volatile aromatic substances, mostly varietal [35,36,37]. The products of the complete transformation chain, from grapes to wine, were analysed to determine their composition, which varies due to the transformation of aroma precursors from yeast during alcoholic fermentation [38].

To confirm the adopted methods’ validity, the soils, fruit, musts and wines’ mineral fingerprints were also assessed to identify vineyards’ mineral markers. Wines typically contain macro-elements (e.g., Na, K, Mg and Ca), micro-elements (e.g., Fe, Cu, Zn, Mn and Pb) and ultra-microelements (e.g., Cr, As, Cd and Ni) [39]. Trace elements less affected by the winemaking process are alkaline earth metals, among which Li, Sr and Rb are the most relevant for geographical origin identification [40]. This additional analysis is useful when certifying a wine’s precise geographical traceability, when testifying to the possible existence of direct markers between the vineyard and its products and when confirming the correctness of the basic assumptions in this study on the ecophysiological adaptation of the vine in response to the abiotic stresses of the vineyard where it grows [41,42,43,44].

This multidisciplinary research is designed and coordinated to involve geological, climatic, agri-food, oenological, geochemical and biochemical analyses. This applied research selects a specific terroir’s component factors and collects essential elements for the interdisciplinary investigation of complex alpine terroirs. An initial summary of the surveys, analyses and measurements is listed in

Table 1. Summary of the surveys and measurements of a terroir’s component factors adopted in the research.

Terroir’ Component Factor	Survey and Analysis	Measurements
Geology and vineyards’ geopedology (VGI method)	Geological geomorphological survey	Recognising sedimentary domains and processes in each vineyard, landscape study
	Soil and/or subsoil	In situ stratigraphic measurement and sediment sampling of each vineyard’s stratigraphic unit/process
Topoclimate (SRI method)	Vineyard’s solar radiation identity	SRI method, in situ solar radiation measurements
Microclimate	Multi-parameter measuring stations: air and soil in situ measurements	Air temperature, dew point, vapour pressure, humidity, rainfall, leaf wetness, wind speed and direction, soil and subsoil salinity, temperature and humidity
Soil sediments analysis	Laboratory tests: qualitative and quantitative analysis	GEO: mineralogy, petrography, particle size and morphological features, mineral composition, geochemistry and quantitative elementary analysis
		AGRO: pH, texture, organic matter, cation exchange capacity, assimilable and exchangeable minerals, heavy metals, trace elements and isotope
Viticulture and wine making	Vineyard and winery practices	Kept constant for all vineyards and wines
Grape variety (Gewürztraminer)	Grape, must and wine targeted oenological, chemical quali–quantitative analyses	Basic chemical–physical parameters, mineral profile, amine profile, total polyphenols, flavonoid polyphenols and aroma profile, varietal compounds in free and bound forms, mineral profile, trace elements and isotope
	Grape, must and wine untargeted oenological chemical quali–quantitative analysis	Bidimensional gas chromatography–time-of-flight mass spectrometry, liquid chromatography coupled to electrochemical detection, near-infrared spectrometry, untargeted volatilome profiles and sensory evaluation [45]

GEO = geological soil sediment testing; AGRO = agronomic soil testing.

and explained in detail in the subsequent sections.

2.1. Geographical Location

This study’s geographical focus area was the South Tyrol wine-growing region located in northern Italy within the Alps, which is a vast geographically varied and challenging environment. Specifically, this wine-growing area is located slightly south of Bolzano in the Bassa Atesina wine sub-region, within a small portion of territory comprising only 6 sq. km around the village of Tramin (Figure 1). Terroir analysis was performed on seven vineyards, identified in this study as vineyards V1–V7. As the basic premise of the study was to maintain a unique winery’s viticultural practice, the wines of the seven plots were made in the same winery. These vineyards were chosen based on the winemaker’s experience, who observed different quality and style expressions in their wines over the last decades.

2.2. Geological and Geopedological Settings

The geological history of the Alps presents all the geographical features characterising wine landscapes today, along with many natural ecological elements that affect plants’ phenological behaviour. Glaciers and rivers transported sediments over hundreds of kilometres and, together with debris flows, shaped these morphological landforms with surficial bodies of sediment. The Alps contain many different parent rocks and minerals, as well as a wide range of more recent sediments from the Quaternary period, which were deposited next to each other or on top of each other during the last millennia. Thus, the area’s geological history has influenced the soils’ properties, shaping their physical structure, mineral composition and acidity.

In the Alpine region, the high spatio-temporal variability in which the soil’s parent materials were formed defines the need to develop a geopedological reference model separate from the more classical soil classification. Classifying a vineyard’s terroir requires knowledge of the area’s geological history, including morphological landforms (e.g., altitude, orientation, slope and insolation) and the geo-mineral, geochemical and physical compositions of soils. Sediments that form Alpine soils have undergone various processes, which mixed the parent rocks’ minerals during several sedimentary cycles and deposited them in often complex ways. Ferretti [26] developed a rational geological model to facilitate a regional terroir analysis of the vineyards. Increasing degrees of detail are applied, starting from regional geology to the physical–mineralogical classification of sediments and vineyard soil.

From a broader geographical perspective, we distinguished the Alto Adige DOC wine region into four main districts and families of parent rocks: metamorphic, igneous (volcanic or plutonic), ancient sedimentary, and dolomitic. The metamorphic and igneous rock families are located in the north and central portions of South Tyrol, respectively, comprising predominantly silicate and quartz minerals and yielding more sandy, stable and acidic soils. These two parent rock families are also the primary sources of the most represented soils’ clay minerals in the wine region, the so-called mixed-layer clays, which play a significant role in the fertility of local soils [29]. Ancient Permian sedimentary rocks can be found only along a thin belt in the south-central portion of South Tyrol, including Tramin. They are important to some vine cultivars because they form original fine-textured soils, with a significant percentage of mixed-layer clays coloured by iron oxides and calcium carbonate, resulting in moderately alkaline fertile soil. In the southernmost headland of the region lies a sequence of dolomite rock formations. The dolomite mineral strongly characterises the vineyards and naturally reduces the soil’s acidity.

To introduce a rational and scientific comparison of Alpine geographical areas and to better understand possible vineyard–soil combinations, the parent rocks, which represent the soil minerals’ source, and all the most recent cycles of erosion, transport and deposition of soil sediments should be considered. Soil sediment formation complexity comprises many interacting processes that proceed in multi-steps, forming parent materials for soil genesis and defining their mineral and physical features.

Table 2. The chemical–physical complexity of the sediments that shape vineyard soils in South Tyrol depends on their geological history and the sedimentary cycles they have undergone during the Quaternary period. High-resolution event stratigraphy can be helpful in a local terroir study.

	1st-Cycle Sediments	2nd-Cycle Sediments	3rd-Cycle Sediments
Formation periods	Throughout the Quaternary period	Glacial periods > 14,000 years B.P.	Last millennia to recent times
Parent material	Rocks	Rocks + remixed minerals of 1st cycle sediments	Rocks + remixed minerals of 1st- and 2nd-cycle sediments
Soil sediments’ parent minerals	Pure mineral selection from local parent rocks	Mixed minerals from local and distant parent rocks or sediments	Remixed materials and minerals from 2nd- cycle sediments and occasionally from local parent rocks
Soil’s sediment sources	Micro-catchment basin near the parent rocks	Meso- and macro-catchment basin	Micro- and meso-basins; macro-catchment basin for alluvium
Main sedimentary process	Weathering, debris, landslides, flash floods, debris flows, colluvial, eluvial	Glacially derived; +1st-cycle processes during the glacial periods; +paleo-fluvial and -lacustrine	1st-and 2nd-cycle processes; +fluvial and different water–sediment flow in mountain torrents; landslides, colluvial and other gravity-driven processes

and Figure 2 summarise the local event stratigraphy in the potential cycle sequence of the latest sedimentary processes involved in the formation of local soil sediments. High-resolution event stratigraphy can exceed the resolution of other systems of stratigraphy [46], such as those adopted in the geological map in Figure 3. The more cycles or events that loose natural sediment has passed through, the more complex the soils’ geopedology. At the same time, more remarkable originality is present in the terroirs of the different vineyards.

The different geographical subzones of the Alto Adige DOC wine region have specific mineralogical combinations in soils, characterised by preferential associations of sediments. Tramin vineyards are located in one of the most complex geopedological situations in the wine region, which can be conceptually identified in the T zones in Figure 2. The soils in vineyards can have all possible cases of sediment cycles and remixed minerals with components from igneous, metamorphic, ancient sedimentary and dolomitic parent rocks.

About 12 different rock Formations and four sedimentary units can be found in the small area around Tramin (Figure 3). The sediments of the illustrated sedimentary structural units are composed of minerals from further metamorphic parent rocks brought from afar by glaciers and the Adige River. A high-resolution geopedological study of vineyards’ soils can be conducted using the VGI method, analysing and interpreting the sedimentary minerals’ origin event, as schematically illustrated in Table 2. This method can help recognise their history and sedimentary processes to distinguish each vineyard’s soil sediments and map homogeneous geopedological zones.

2.3. Mineralogy of Soil Sediments

The parent material’s constituent minerals determine soil fertility [47], which, in combination with its texture, can help affect grape quality and indirectly influence a wine’s organoleptic properties [48]. To recognise and classify this factor, several analyses of soil mineral composition were conducted. These tests can provide precise information on the soil’s different parent rocks and, its source geographical areas, helping to define the parent material’s sedimentary origin and frame any soil’s geopedological identity.

To determine the crystalline phases, X-ray diffraction (XRD) of 22 soil samples were performed. Both soil and subsoil samples were tested to study the sediment’s heterogeneity and anisotropy in the whole layer affecting the vine root system (vine roots were found up to 2 m deep). Samples were collected at depths of 20–155 cm. The morphological features were determined through optical microscopy and quantitative elementary analysis using scanning electron microscopy–energy dispersed spectroscopy (SEM-EDS).

Powder diffraction analysis (XRD) allows for qualitative and quantitative estimations of the soil’s crystalline mineral phases, and it also recognises parent rocks’ minerals from various Alpine geographic areas and the presence of neogenic minerals. These tests on the soil’s inorganic component help classify each parent material’s history and sedimentary process as well as soil fertility predisposition.

An X-ray fluorescence spectrometry (XRF) analysis was also conducted to determine the chemical composition in terms of the major elements present in the soil samples. The major chemical elements detected, expressed as a percentage of their oxide in the total sample, were SiO₂, TiO₂, Al₂O₃, Fe₂O₃, MnO, MgO, CaO, Na₂O, K₂O and P₂O₅. The amount of volatile elements (H₂O and CO₂) in hydrated minerals and carbonates was determined through calcination at 1000 °C, and as part of the chemical composition, it was expressed loss on ignition (LOI).

2.4. Agrochemical and Physical Soil Analyses

Agronomic analyses were conducted on 23 soil and subsoil samples from the seven vineyards to gauge the pH (using the water ratio 1:2.5 potentiometric method); the texture according to the U.S. Department of Agriculture (USDA), organic matter (OM) (through elemental analysis of organic carbon) and cation exchange capacity (CEC) (titration in barium chloride and triethanolamine); qualitative and quantitative soil testing for assimilable and exchangeable minerals, and ethylenediaminetetraacetic acid (EDTA) diethylenetriaminepentaacetic acid (DTPA) agents for heavy metals. Furthermore, trace elements in soils were measured using a multi-element analytical method known as inductively coupled plasma–mass spectrometry (ICP-MS) [49].

Geotechnical grain size analyses of all soil samples (grain size < 10 cm) were conducted according to DIN ISO 17892-4 standards [50]. Unlike the soil texture test based on the aforementioned USDA system, which measures the 0–2 mm material fraction, this geotechnical test provides more complete and significant data on the parent material. In our opinion, in many terroir studies, a full-range granulometric analysis is essential in investigating each vineyard’s geopedological identity, allowing for the evaluation of sediments’ depositional processes and their agronomic and geotechnical behaviour, thereby aiding in the classification of soil types of different origins.

2.5. Local Climate

In an attempt to reduce climatic variability, this terroir research focused on a small wine-growing area around Tramin. In an area of approximately 6 km² stretching in a north-south direction (

Table 3. Position of the environmental air and soil monitoring stations used in this study.

Locality	Vineyard	Location	Height m a.s.l.	SRI Index *
Söll	V1	46°21′40.2″ N 11°14′35.0″ E	429	78.6
Am Sand	None	46°20′41.1″ N 11°15′06.8″ E	235	76.3
Plon	V7	46°19′54.3″ N 11°14′03.9″ E	339	68.2

(*) SRI = solar radiation identity index (See sub-Section 2.7).

), three environmental monitoring stations that collected environmental information during the 2019 wine season were installed approximately 2 km apart. The stations for vineyards V1 and V7 were installed at the north and south extremes. The Am Sand station was placed between the two, but at a lower altitude and outside of all the tested vineyards. The parameters measured were solar radiation, air temperature, relative humidity, dew point (vapour to liquid), vapour pressure (liquid to vapour), precipitation, leaf wetness and wind speed and direction. In addition, soil and subsoil temperature, salinity and humidity were measured at depths of 10–120 cm.

2.6. Soil Moisture and Temperature

The soil temperature in the root zone acts as an abiotic stress and affects the biochemical processes in the soil and the nutrition of the plant. Soil temperature depends on several factors: water content, energy balance (soil mineralogy, colour and soil cover), vineyard exposure, altitude and topography. Rapid temperature fluctuations can be due to a convective exchange with air and rainwater. The soil temperature profile and humidity were measured continuously at a depth of 10–1.2 m with 12 sensors. Soil moisture sensors determined the volumetric water content by measuring the soil’s dielectric constant using capacitance technology and soil temperature (Sentek Drill & Drop TriSCAN device. Pessl Instruments Gmbh, Weiz, Austria).

2.7. Vineyard SRI Index

The SRI index is a new approach to applied research and decision-making in the viticultural sector, implemented to classify vineyards’ topoclimates in an accurate and comparable manner [26]. It serves to distinguish and synthesise the complex environmental conditions of vineyards in mountainous areas, particularly their topoclimates. This factor includes numerous independent variables linked to the vineyard, such as slope, orientation, exposure, the position of the sun, zenith on the horizon, and mountain shadow. The SRI assigns an identity value to each vineyard, i.e., to each possible homogenous wine lot, which takes into account all these sub-factors. Therefore, the index rate, which is linked to the biosynthetic activity of the vines, can be related to the quality of the vineyard’s grapes and wine. Ferretti [7] found that the topoclimate is represented adequately and accurately using the SRI index, allowing for ripening incidence to be estimated quantitatively. For example, a higher must sugar content of ca. 0.8 °KMW (Babo) corresponds to an increase in SRI of 10 points. A fairly evident effect on the wine’s quality parameters can be appreciated with differences in the SRI index of more than 5–10 points.

2.8. Grape, Must and Wine Analyses

Products from the 2019 harvest of the Gewürztraminer cultivar from the seven different vineyards, totaling 21 grapes (harvested on 3 and 4 October 2019), must and wine samples, were analysed. The targeted oenological chemical analysis was performed at the same time point and included the basic chemical–physical parameters, mineral profile (ICP MS-ICP OES), amine profile (high-performance liquid chromatography method), total polyphenols (spectrophotometric method), flavonoid polyphenols (profile), spectrophotometric method and aroma profile, including varietal compounds in free and bound forms (gas chromatography-mass spectrometry method). Oenological chemistry analysis followed the official methods of the International Organisation of Vine and Wine and the European Union regulations, as listed in [51]. Based on the analytical results, this study focused mainly on the varietal volatile aromatic substances characterising the Gewürztraminer variety [52,53,54]. Aspects related to the technological choices made in the vineyard and cellar were not evaluated because they were kept constant for all the vineyards studied.

2.9. Statistical Analysis

Most of the variables were numerical, and the main descriptive statistics (means and standard deviations) of these variables were determined and reported. Only numerosity was determined for the ordinal variables. As the units were few (n < 10), it was considered appropriate to study the distributions of the variables, the normal probability plot for assessing whether a dataset is normally distributed and the presence of outliers. Visually, boxplots were created; numerically, the Grubbs test (G) and Dixon test (Q) were conducted. The Dixon test is non-parametric, and it was conducted because of the small number of units.

A local outlier factor (LOF) algorithm was carried out. It is an unsupervised (or semi-supervised) machine learning algorithm that computes a local density for observations with user-given k-nearest neighbours: A normal data point has an LOF of 1.0–1.5, while anomalous observations have a much higher LOF. The higher the LOF, the more likely it is to be an outlier.

Between-type analyses were not conducted because the aspects detected were different from each other, the number was too low, and they were not considered of interest in this work.

3. Results

3.1. Air Temperature

Data from the samples showed consistent seasonal temperature trends for all three environmental measuring stations. The data comparison between V1 and V7 confirmed a reduction in air temperature at an altitude of approximately 0.9 °C per 100 m, matching the typical regional values during summer and indicating temperature decreases of 0.6–1.0 °C for every 100 m in altitude [26] Temperature differences were registered in the vineyards during the vegetative period and were attributed to the mesoclimate conditions of this mountain environment, especially local topographical settings. Daily daytime maximum and nighttime minimum temperature changes reached approximately ΔT 6 °C between the three stations. Differences were found between minimum temperatures of down to 3.5 °C and between maximum temperatures of up to 1.5 °C. Higher diurnal temperature variations were measured along the narrow Höllental torrent Valley (Figure 4), which descends directly from the dolomite mountain peak.

3.2. Rainfall

Monitoring was conducted from 23 March to 31 October 2019. During this period, 78 rainfall episodes were recorded, which, compared to those at the three monitoring stations, were simultaneous and of the same intensity. In some sporadic events, slightly different rainfall was observed at the stations. This finding supports the fact that mountainous areas are favourable to local thunderstorms of varying intensities. The total amount of precipitation during the March–October period was similar: V1 = 686 mm, V7 = 683 mm and Am Sand = 664 mm.

3.3. Soil Moisture

The changes in volumetric water content measured at the V1 and V7 stations accurately followed the rain events. V1 soil was 4–6% wetter than V7 soil, but wetness was higher close to the surface because of the lower evapotranspiration rate due to a lower permeability. Starting from about −40 cm deep, the moisture remained more stable over time. V7 was wetter, and the humidity level varied only seasonally. In the subsoil, the highest value was 30% in winter, and the lowest value was 18% in late summer. Moisture increased again after summer with autumn rains and frost. Throughout the growing season, the subsoil in V7 was regularly 8% wetter than that in V1, and the highest available moisture-holding capacity was shown by V7 loamy soil compared to the more clayey soil (S. §3.5 ‘Soil textures’). More energy is needed to heat soil with a higher water content. This can explain the data measured in V7, where the soil temperature was lower due to its higher heat capacity.

3.4. Soil Temperature

Air temperature did not have an immediate effect on soil temperature. It had a certain inertia and was appreciated only on a monthly and seasonal scale (Figure 5). Conversely, thermal excursion in more superficial soils depends on solar radiation, which acts quickly. During the vegetative period, the first 30 cm of soil near the surface was warmer, and the daily temperature trend was cyclical, strictly linked to solar energy. Vineyard V1 had a warmer surface soil, even if it was 100 m higher than that of V7, and the air temperature was 1° C colder in V7 than in V1. Specifically, the soil at a depth of 10 cm in V1 was 1 °C warmer than in V7. This confirms the topoclimate SRI values factored into the study, as the SRI index of V1 was about 10 points higher than that of V7. Deep down in the soil, solar radiation exerted no effect on temperature. Figure 5 shows that soil temperature depends not only on air temperature and the rate of solar radiation on the ground but also on texture and humidity. With the same energy rate, wetter soils heat up more slowly. As noted above, the studied soils had different fine textures and moisture levels, with V7 being wetter than V1. Thus, the V1 soil became warmer more easily in summer and colder more easily in autumn. In the following paragraphs, the different mineral characteristics of the soil are illustrated. These factors lead to varying physical behaviours, as shown in Figure 5, which compares subsoils from vineyards V1 to V7.

3.5. Soil Texture

An analysis of the ternary diagram proposed by the USDA (Figure 6a) based on a particle size test of the medium-fine fraction (<2 mm) found that the soils contained a low clay fraction and varied from silty loam to sandy loam. Most soils were characterised by an intermediate loam composition. This texture test was only partial, as it did not analyse the soil skeleton’s gravel fraction. It favoured the clay fraction and thus did not provide sufficient geopedological information to distinguish between soils in the Alpine context, in which sediments predominantly comprise coarse gravelly fractions compared to the fine clay fraction. In our case, we found that the soil gravel fraction was 41.1%, on average, and the clay fraction was only 3.7%.

To better determine the geological and geopedological processes that led to soil formation and to distinguish the vineyards’ terroirs, texture tests on the entire soil sediment were conducted. The particle size distribution (PSD) of soil can markedly affect terroir research, as a granulometric distribution analysis from studying sediment sorting, uniformity and curvature coefficients and skewness provides information on the depositional environment, soil transport energy, deposition agents and sedimentary processes. In this case study, a well-textured sorting indicates the laying down of material in a water-governed environment (e.g., V1 and V2). Conversely, poor sorting with a coarse grain size indicates a torrential environment, which is a debris-flow-type process (e.g., V4) or the presence of debris particles from nearby rocky outcrops (e.g., V3) in our case. Discontinuities in the grain size curve indicate the mixing of sediments of different origins (V5–V7). Figure 6b illustrates how the samples from vineyards V1 and V4 are positioned on opposite sides of the triangle and differ significantly from each other.

3.6. Soil and Subsoil Mineralogy

3.6.1. Mineralogical Analysis Using XRD

XRD and quantitative phase analysis using the Rietveld refining method were employed to determine the crystalline phases in the soil samples, as illustrated in

Table 4. Weight fractions (wt%) of the crystalline phases identified by Rietveld refinements in the 22 soil and subsoil samples.

Mineral Phase	V1-2	V1-3	V1-4	V1-5	V1-6	V2-2	V3-1	V3-2	V3-3	V4-1	V4-3	V4-4	V5-1	V5-2	V5-3	V5-4	V6-1	V6-2	V6-3	V7-1	V7-2	V7-3
Quartz	36.7	30	32.1	38.9	34.1	21.8	35.4	38.9	30.4	19.3	14.6	31.9	29.8	24.4	29.9	31.6	24.9	19.4	21.4	38.9	33.1	40.8
Muscovite	14.4	19.9	20.9	20.8	14.4	5.4	16	13.8	17.3	3.6	4.3	2.5	7.1	5.6	4.4	8.3	7.9	4.5	5.4	6.3	7.1	7.8
Dolomite	20.1	18.6	12.5	12.6	11.6	51.4	16.6	16.2	23.1	54.8	51.9	43.3	37.5	43.6	35.1	32.7	40.4	52.3	36.1	14.6	17.2	16.6
Kaolinite	9.1	11	9.9	8.4	7.7	1.04	9.8	7.2	6.9	2.8	4.3	2.3	4.5	2.8	3.5	4	2.97	1.53	2.63	4	3.85	4.2
Albite	3.5	2.1	1.2	3	1.6	0.9	1.5	1.2	2.1	1.85	2.56	4.8	2.2	3.2	3.2	4.2	1.7	1	1.28	0.9	0.98	1.2
Calcite	2.94	5.56	7.74	0.61	8.6	7.38	4.09	8.15	6	5.87	8.21	1.33	1.1	2.18	1.53	0.75	4.21	12.8	14.8	17.2	20.4	10.4
Orthoclase	5.29	3.06	3.48	5.1	4.8	3.99	3.47	2.66	3.81	2.6	1.93	3.46	2.83	3.2	2.89	3.8	5.2	3.31	4.4	5.22	4.6	5.1
Hematite	0.34	0.72	0.36	0.26	0.18	0.25	0.29	0.21	0.38	0.94	0.1	0.03	0.11	0.23	0.01	0.06	0.34	0.33	0.37	0.47	0.49	0.27
Clinochlore	0.69	1.62	1.28	2.9	0.93	2	3.1	2	0.8	3	5.6	1.64	3.9	5.1	5.7	3.2	1.19	0.7	0.91	1.15	0.66	0.73
Hornblende				0.15							0		0.78	0.53	0.21	0.56			0.6
Clay mixed layer	7	7.4	10.6	7.3	16.1	5.8	9.8	9.7	9.1	5.3	6.6	8.8	10.1	9.2	13.6	10.7	11.2	4.1	12.3	11.1	11.6	13

.

Triangular plots were constructed to examine the results of the quantitative XRD analyses (Figure 7). The mineral phases that best define the depositional processes typical of the valley side where the vineyards are located were detected and evaluated. This helped recognise the parent rock that largely influenced the nature of the resulting soil (cf. Material and methods §2. siliceous, dolomitic or terrigenous) and the sedimentary process that formed the soil. The minerals were subdivided according to their petrographic origin and agronomical importance into carbonate (dolomite and/or calcite), silicate and clay minerals (phyllosilicates). Figure 7a highlights vineyards V1 and V3, with a predominantly siliciclastic component (quartz + feldspars + phyllosilicates) and a lower carbonate fraction. Vineyard V7 has an important siliciclastic component and shows a partial increase in calcite minerals. Found in the south of Tramin, this mineral represents a characteristic marker mineral from rocks outcropping only in the studied area’s immediate proximity. In the second group of vineyards (V2 and V4–V7), which was less delimiting than the previous group, the carbonate mineral composition (dolomite + calcite) became progressively larger.

The weighted analysis of these initial mineralogical data, excluding the outlier calcite mineral, distinguished three homogeneous groups of soils more clearly, as they were associated with specific parent rocks and sedimentary processes (Figure 7b). It is now easier to recognise the analysed vineyards’ varying soil types. For example, the calcite mineral’s pedological origin and the related soils’ geographical location were identified more precisely.

3.6.2. Chemical Analysis Using XRF Spectrophotometry

The detected features with the most variation in the soil samples were SiO₂, Al₂O₃, (MgO + CaO) and LOI procedure results. They highlighted the varying ratios between silicate and carbonate minerals [55] as well as in the small examined area. The analyses found that the samples from vineyards V1, V3 and V7 were characterised by abundant silica (32–56%) and aluminium (9–17%) content, a siliciclastic quartz–clay component and a low-to-moderate calcium + magnesium content (11–31%). Regarding the latter, the analysis indicated that the carbonates in soils were of dolomitic origin, with only V7 registering higher calcite contamination. Therefore, substantial mineralogical differences were observed in the vineyards’ soils, indicating varying geopedological compositions.

Although the XRF test results were useful in distinguishing the soils’ nature in general, they did not conclusively define the soils’ complex mineralogy in this study. VGI research has indicated that XRD tests are more efficient in chemical–mineralogical analyses.

3.6.3. ICP-MS Elemental Analysis of Soils

The ICP-MS agronomic analysis of the soil mineral elements also distinguished the different geopedological conditions in the vineyards. Some chemical elements can act as characteristic mineral markers (Figure 8) and provide information on the geopedological differences between vineyards and the elemental mineral composition released through mineral weathering. Furthermore, when used as a petrochemical indicator together with data from mineralogical and geological analyses, elemental analysis helps trace soils’ origins and aids in the creation of more precise soil maps.

The analysis of the Rb/Sr ratio helped assess whether any parental link existed between parent rocks and soil-forming minerals in the studied vineyards (Figure 9). For example, regarding vineyards V5 and V6, the geological origins of the so-called local “mixed soils” were indefinite. Vineyard V5’s high Rb/Sr ratio indicates that its soil contains minerals from distant metamorphic rocks present only in the far north of Tramin. However, vineyard V6’s low Rb/Sr ratio indicates an origin from closer rocks, that is, the high dolomite mountains on the valley side near the village of Tramin.

3.7. VGI

Table 5. VGI classification for the seven vineyards.

Vineyard	Texture Class		Permability	pH	SRI	Altitude (m a.s.l.)	Parent Rock (Prevalent)	Sedimentary Process	Domain	Ground Water	Soil Mineralogy **	Mineral Markers	Chemical Markers
	NRCS-USDA	Sheppard	k (m/sec) *								(%)
1	loam	sandy silt	1.2E-01 ÷ 6.3E-01	8.1	78.61	435.0	Metamorphic and ancient sedimentary	Undifferentiated till deposit	Siliciclastic	no	Qz (34), Do (15), Ca (5), Phs (38)	Kaolinite Hematite	Be, B, Rb, Ag, Pb
2	silt loam	Silt-gavelly silt	3.7E-02 ÷ 8.6E-02	8.4	75.50	222.4	Dolomite	Mixed fan: stream flow	Dolomitic	no	Do (51), Qz (22), Ca (7), Phs (14)	Dolomite	Ca, CaCO3 (t), Cu
3	loam	loam	1.0E-01 ÷ 2.5E-01	8.3	79.02	436.3	Metamorphic and ancient sedimentary	Glacial lodgment till deposit	Siliciclastic	no	Qz (35), Do (19), Ca (6), Phs (35)	Kaolinite Hematite	B
4	sandy loam	sandy gravel	2.5E-01 ÷ 4.0	8.4	76.47	361.7	Dolomite	Ancient mixed fan deposit	Dolomitic	no	Do (50), Qz (22), Ca (5), Phs (17)	Dolomite Hematite	Ca, CaCO3 (t)
5	loam	silty gravel	1.4E-02 ÷ 1.2E-01	8.2	78.29	315.3	Dolomite	Ancient mixed fan deposit	Mixed	no	Do (37), Qz (29), Ca (1), Phs (25)	Hornblende Clinochlore	Ag, Mg
6	loam	silty gravel	4.8E-02 ÷ 2.2	8.2	72.65	323.0	Dolomite and ancient sedimentary	Mixed fan: stream and debris flow	Mixed	no	Do (43), Qz (22), Ca (11), Phs (18)		Ca, CaCO3 (t), Cu, Zn, S
7	silt loam	loam	5.7E-01 ÷ 6.6E-01	8.6	68.16	346.3	Ancient sedimentary	Gravity driven deposit	Terrigenous	no	Qz (38), Do (16), Ca (16), Phs (24)	Clay mixed layerHematite, Calcite	B, Ca, CaCO3 (a), Sr

* according to [56]. ** Prevalent minerals: Qz, quartz; Do, dolomite; Ca, calcite; Phs, phyllosilicates; ^(t), total carbonate; ^(a), active carbonate.

presents the VGI geopedological classification of the studied vineyards’ soil sediments and the geological information essential for the terroir investigation. VGI data accurately describe the origins of the geopedological non-homogeneity between the studied vineyards, which is primarily due to the complex and varied geological evolution of this small portion of land around Tramin. From a geopedological perspective, VGI also distinguishes and qualifies soils’ varying chemical–physical conditions, which can indicate some abiotic stress circumstances for vines.

Information about parent materials’ physical and chemical features with practical implications for wine growing is shown in Table 5. For example, the macro-differences in the carbonate mineral component (i.e., the presence of calcite) explain V7’s higher pH levels. The sedimentary process marks the textural and mineralogical features affecting the highly varying moisture and oxygen contents in V1 and V4. Clay fraction and mineral oxide amounts lead to different soil fertility conditions.

The data in Table 5 indicate a higher amount of CaCO₃ in Vineyard V7, which is connected to calcite, marking the active carbonate amount and the highest pH range for this soil. This calcite mineral is related to a specific rock (Contrin Formation), which outcrops directly above vineyard V7 and whose calcite was set in place through gravity-driven sedimentary processes. Vineyard V1 has fine-textured soil, mostly containing siliciclastic glacial till sediments and a noteworthy amount of phyllosilicates, clay minerals and iron oxides. V1′s parent material originates from the entrainment of minerals from distant metamorphic parent rocks due to the Adige glacier’s movement. The parent material was deposited at some distance down-ice, with some influences from a lacustrine environment. Vineyard V4 contains dolomite parent minerals deposited by the Höllental torrent (Höllentalbach), forming an ancient mixed fan about 33,000 years B.P. [57]. This debris flow sediment has a coarse texture with a high total carbonate content, but it is more equilibrated in active carbonate. In reality, the rich presence of dolomite naturally reduces soil acidity. However, unlike limestone, which is a close relative of dolomite, this rock is less soluble, and the structure of its soils remains more stable and porous over time.

The VGI data set listed in Table 5 facilitates the geographical definition and perimeter delineation of homogeneous geopedological domains in which similar soil conditions are repeated.

As listed in the right-hand columns of the table, the markers of mineral and chemical elements have been recognised in soils (S. §7.3 ‘ICP-MS elemental analysis of soils’). VGI research has confirmed that some of these chemical elements can be found in musts and wines and may be used to trace the geographical origins of food products [40,41,43].

3.8. SRI Index

The SRI values of the seven vineyards ranged between 68 and 79 points (Table 5) due to differing topoclimate situations and mountain shade conditions. All vine rows were oriented towards east-southeast, but there were differences in steepness and were covered differently day-to-day by afternoon shade from the mountains. The SRI index of vineyards V1 and V7 was confirmed by solar irradiance measured between May and October 2019 [7] at the two weather stations, where data were sampled continually on site using a pyranometer.

3.9. Chemical and Oenological Analyses

3.9.1. Statistical Analysis Results

The results alongside the manuscript can be found in the Supplementary Materials section in Tables S1–S5. The wine dataset consisted of seven wines, from which 32 main characteristics were found. The means and standard deviations of all the numerical variables, divided by type are reported in

Table 6. Means and standard deviations of mineral elements in wine (see Table S2 in the Supplementary Materials).

Variable	Unit	Mean	Standard Deviation
Barium	µg/L	37.51	16.57
Beryllium	µg/L	0.19	0.10
Boron	mg/L	4.84	0.41
Calcium	mg/L	54.56	8.31
Iron	mg/L	0.14	0.05
Manganese	mg/L	0.71	0.36
Potassium	G/L	1.34	0.17
Copper	mg/L	0.13	0.06
Rubidium	µg/L	0.64	0.25
Sodium	mg/L	16.87	2.63
Strontium	µg/L	0.14	0.04
Zinc	mg/L	0.51	0.14

,

Table 7. Means and standard deviations of amino acids in grape (see Table S3 in the Supplementary Materials).

Variable	Unit	Mean	Standard Deviation
L-Alanine	mg/kg	97.21	67.57
L-Arginine	mg/kg	400.50	266.49
L-Glutamine	mg/kg	250.40	157.42
L-Methionine	mg/kg	1.74	2.83
L-Serine	mg/kg	62.07	22.15
L-Threonine	mg/kg	92.77	64.52
L-Tyrosine	mg/kg	5.56	1.71

and

Table 8. Means and standard deviations of mineral elements in wine (see Table S4 in the Supplementary Materials).

Variable	Unit	Mean	Standard Deviation
4-Vinylphenol	mg/L	1.77	0.87
4-Vinyl Guaiacol	mg/L	6.29	2.34
ß-Phenylethyl	mg/L	20.49	4.04
Rose oxide tot.	mg/L	0.009	0.002
3-Mercaptohexanol	mg/L	0.54	0.18
Citronellol (free-form)	mg/L	0.10	0.03
Citronellol (bound-form)	mg/L	0.03	0.01
Citronellol tot (free + bound)	mg/L	0.13	0.02
Geraniol	mg/L	0.71	0.44

.

Data on the measured mineral elements are presented in Table S2. Silver, cerium, lanthanum and lead reported the same value for all wines: <0.01 (silver), <0.03 (cerium and lanthanum) and <16 (lead).

Regarding the outliers, the boxplots of all the numerical variables of interest were first reported (with only four in terms of Ba, Be, Mn and Rb) (Figure 10 and Figure 11).

For the mineral elements, we focused only on the most important for the research and those that were not affected by agricultural practices.

As shown in Figure 10, outliers were found in the beryllium and rubidium distributions. An outlier was found in each amino acid (Figure 11). For the polyphenols (Figure 12) outliers were found in 4-vinyl guaiacol, geraniol and 3-mercaptohexanol.

To determine whether these outliers were important, the Grubbs test was conducted for each case. The results are shown in

Table 9. Grubbs and Dixon tests.

Variable	Grubbs Test Statistic (p-Value)	Dixon Test Statistic (p-Value)
Beryllium	1.755 (0.144)	0.4375 (0.194)
Rubidium	2.094 (0.010)	0.662 (0.0136)
L-Alanine	2.242 (<0.001)	0.830 (p < 0.001)
L-Arginine	1.923 (0.056)	0.453 (0.169)
L-Glutamine	2.093 (0.011)	0.592 (0.037)
L-Methionine	2.249 (<0.001)	0.888 (p < 0.001)
L-Serine	2.123 (0.007)	0.629 (0.022)
L-Threonine	2.166 (0.003)	0.740 (p < 0.001)
L-Tyrosine	2.069 (0.015)	0.596 (0.035)
4-Vinyl Guaiacol	2.027 (0.023)	0.504 (0.103)
Geraniol	0.099 (0.048)	0.310 (0.501)
3-Mercaptohexanol	3.289 (0.023)	0.415 (0.234)

.

The two tests did not always agree on the detection of outliers. For the Grubbs test, the values detected graphically as outliers were always verified as such (p < 0.05), except for beryllium and L-arginine variables. For the Dixon test, only the outliers of rubidium and amino acids (except L-arginine) were considered.

The cases in which some values did not result in insignificant outliers at the test level were due to the fact that they were not very far from the rest of the distribution (as shown graphically).

The results of the LOF analysis are reported for each variable, with possible outliers in

Table 10. Grubbs and Dixon tests.

Variable\Units	V1	V2	V3	V4	V5	V6	V7
Beryllium	7.00	1.17	1.00	1.00	1.00	1.00	1.00
Rubidium	12.25	1.00	1.00	1.33	1.33	1.00	1.00
L-Alanine	11.97	1.02	1.02	1.32	1.00	1.02	1.00
L-Arginine	2.07	1.00	1.00	1.01	1.06	1.00	1.00
L-Glutamine	5.89	1.03	1.07	2.83	1.01	1.02	1.07
L-Methionine	23.67	1.00	1.00	1.00	1.00	1.98	2.37
L-Serine	3.24	1.00	1.00	1.89	1.67	1.00	1.00
L-Threonine	4.05	1.01	1.00	1.09	1.00	1.00	1.00
L-Tyrosine	6.39	1.00	1.39	1.00	1.37	1.11	1.39
4-Vinyl Guaiacol	5.95	2.97	1.00	2.03	1.15	1.00	1.00
Geraniol	1.00	1.65	1.11	3.09	2.48	1.00	1.00
3-Mercaptohexanol	1.36	3.79	1.00	2.72	1.07	1.36	1.00

.

The highest scores belonged to the points that became outliers through visual analysis.

Thus, even if only seven units were available for each investigated variable, the values between the wines are so different from each other as to result in a significant outlier. However, that could affect other subsequent statistical analyses if these were carried out. We did not run these analyses because they were beyond the scope of this paper.

3.9.2. Chemical and Oenological Analyses of Grapes, Musts and Wines

The analysis of musts and wines on the varietal most typical aromatic substances of the Gewürztraminer cultivar indicated quantitatively significant differences between the respective vineyards. In particular, more characteristic differences were observed in the products from vineyards V1 and V4. Figure 13 shows that thiol compounds with a tropical scent (3-mercaptohexanol) had 13 times the olfactory threshold and were significantly lower in V4. Conversely, the quantities of free-form geraniol, 4-vinyl guaiacol and citronellol were higher. The phenolic compounds with a spicy note (4-vinylphenol and 4-vinyl guaiacol), which have significantly high amounts from a sensory perspective, characterising the smell of all the samples, were different but more equilibrated in V1. The total rose oxide content (i.e., the sum of isomers in both free and bound forms) indicated a good amount in all the samples, but it was more equilibrated in V1. Among the aromatic alcohols derived from fermentation, significant differences were found in the β-phenylethyl alcohol content, which is characteristic of rose scent, and was less present in V1. Grapes also differed in their total content (free forms + bound forms) of technologically important terpene compounds, such as geraniol and citronellol.

For the total polyphenol content, considerable differences were found in the vineyards. Flavanols are molecules that do not provide sensory notes of the olfactory type, but they contribute to taste perceptions linked to tannicity and are correlated with wine’s ageing process. In particular, V1 showed lower levels of total polyphenols, even in wine.

Variable amounts of the amino acids L-alanine, L-arginine, L-glutamine, L-asparagine, L-serine and L-threonine were measured in the grapes of all the vineyards. Specifically, V1 was an outlier, with a higher amino acid content than the other vineyards (Figure 11). Important organic molecules play a role during natural winemaking processes, optimising and promoting the biochemical transformation of varietal aromatic precursors into secondary and tertiary aromas.

Some natural metallic element content indicated widely varying trends in soils, grapes and musts. In V1′s wine, Ba (+78%), Rb (+82%) and Mn (+57%) were found in significantly higher amounts than the average levels, whereas Be (−90%) was well below the average of other wines (Figure 10).

However, levels of bioavailable minerals in wines were often lower, or at least less evident, than those measured in soils and musts. This indicates that these minerals were actively involved in winemaking processes, interacting with yeast or biochemical processes and separating from the liquid phase. In fact, the comparison between the soils’ geochemical analysis results and the respective chemical–oenological analysis indicates better mineral element traceability in grapes and musts. However, in the wines, these mineral elements occurred in significantly lower quantities, sometimes even below the detection thresholds (e.g., Ag, La, Ce and Pb).

The amounts of the isotopic minerals Rb and Sr also significantly decreased from must to wine but remained highly significant in this study. Given that elements are taken up by roots and transferred to the grape in the same isotopic proportions as those found in the soils [59,60,61], we examined the relationship between Rb/Sr ratios in both soils and oenological products and conducted pairwise correlations using the Pearson method. The high significance of the correlation coefficient (R) was evaluated using Pearson’s p-value at a p < 0.001 significance level. The study revealed that the Rb/Sr isotopic ratios in soils, grapes, musts and wines maintained a constant relationship between their respective vineyards (Figure 14). Therefore, the measured elemental fingerprint confirms the existence of a close link between the local terroir and agri-food products. The isotopic element analysis provided an effective and rare tool for verifying the geographical origin of local wines [62].

4. Discussion

Recent literature has confirmed that advances in measuring the terroir effect on wine quality can be achieved by deconstructing the effects of quantifiable soil and climate factors on grape and wine aroma compounds. In the present study, the effects of the terroir factors on Gewürztraminer wines were assessed through separate analyses and comparisons between the main abiotic factors defining the terroir (temperature, radiation, water and soil geopedology), inducing specific aromatic typicity [31], and the chemical oenological elements most typical of the Gewürztraminer cultivar on grapes, musts and wines [35,63]. A qualitative and quantitative study of the factors comprising the local terroir enabled us to exclude those that were more or less constant between the vineyards. We recognised other specific environmental elements, particularly soil features, that varied significantly and could affect vine ecophysiology and fruit quality differently. Note that technological practices and winemaking were kept strictly constant for all vineyards, as plant material and human management could modulate aromatic expression [64]. All the vineyards have the espalier training system, a similar north–south grapevine orientation, and the same defoliation, fertilisation (no soil fertiliser) and irrigation practices (drip irrigation only in case of water stress). The comparison of the vineyards’ ecological factors and related aroma markers identified interesting correlations: the studied vineyards’ geo-mineral and physical soil conditions in their most extreme configurations (V1clay silicate of glacial origin or V4 sandy dolomitic soils of debris flow origin) were related to varying amounts of amino acids, primary varietal aromas and polyphenols found in grapes, musts and wines. Thiol compounds with a tropical scent, phenolic compounds with spicy notes or terpenic compounds (e.g., geraniol and citronellol) were related differently to fine-textured, siliceous, clayey soils compared to coarser-textured dolomitic soils. The major abiotic drivers of terroir expression examined in this study are discussed below.

4.1. Water Status

Wine aromatic typicity is significantly affected by vine water status [31], which is a major driver of terroir expression [3] that depends on climatic conditions and soil type (e.g., texture and soil water holding capacity) [19]. We analysed both factors. In this study, drip irrigation of vines was scheduled only in rare cases of water stress, which did not occur in the 2019 vintage. The climate monitoring station did not register notable differences in rainfall amounts because of the area’s small extension. However, the moisture measured in the soil was not constant and, together with the mineral–physical soil features, significantly affected soil and subsoil temperatures. According to Duffková [65], loamy soils exhibit the highest available moisture-holding capacity compared to clayey or sandy soils. This is a factor that can significantly affect biological and biochemical soil activities, as water has a high specific caloric capacity (e.g., wet soils warm up more slowly than dry soils) [66].

4.2. Air Temperature

Air temperature is a highly variable and important parameter related to sugar and anthocyanin composition [12,67,68]. Only negligibly small variations in air temperature among the vineyards were measured, with reference to the trends and values of the average daily temperatures (∆T < 1 °C). In reality, no significant differences were observed in the vineyards’ bud break, flowering and technological maturity dates, as all vineyards harvested their grapes within four days of one another in mid-October. Moreover, in accordance with our research, outside of the seven vineyards, we observed an interesting microclimatic situation with a more accentuated daily temperature range along the narrow Höllental Valley that crosses the territory: microclimatic conditions that could affect grape and wine composition [69]. These observations involved very small portions of the territory, which should be monitored carefully in future studies of the Alpine environment, possibly with capillary control of each vineyard’s air temperature.

4.3. Solar Radiation and Soil Temperature

Solar radiation plays a notable role in defining wine anthocyanin concentrations in grape skin [15,70,71]. This class of flavonoids can give Gewürztraminer grapes low-to-moderate colouration. However, it is not a significant element in this study of a white wine grape variety. As solar radiation is also related to the vineyard topoclimate, it elicited more environmental stress analysed in the study: soil temperature in the root zone affects the phenology and timing of ripeness [9]. Radiation was assessed as a topoclimate vineyard element using the SRI method, accompanied in situ by solar radiation and soil temperature measurements. Slightly different topoclimate conditions were found in the vineyards, locally exerting a small effect on the temperature of the most superficial soil. The latter was more sensitive to soil and subsoil mineralogy, textures and the soil’s water content. Indeed, previous literature has found a strong correlation between a soil’s temperature and its physical–mineral characteristics, as temperature is highly affected by soil structure [9] and, as seen above, water content. At the beginning of September, an important and rapid reduction in soil temperature was measured in V1 and remained steadily lower throughout the long pre-harvest period (∆T about −3 °C). A lower temperature in V1′s finer-textured siliceous soils could be a noteworthy physical value and could even have influenced the soils’ biochemical processes, nitrogen supply, plant ecophysiology and metabolism and biosynthesis of certain amino acids and aroma compounds in grapes and wines. This evidence was also confirmed by the winemakers themselves, who noted a certain delay/slowing of vegetative activity in vines growing in finer siliceous soils. They also observed local vines’ aromatic characteristics, detecting a definite stylistic affinity between those from vineyards at lower altitudes with fine “red-coloured” soils and those from vineyards at higher altitudes (>250–300 m) with more dolomitic soils.

4.4. Soil Geopedology

This study examined soils’ abiotic aspects using sophisticated mineralogical analyses (XRD, XRF, thermogravimetric and ICP-MS). The results provide new qualitative and quantitative information to aid in local soil classification. All of the observed vineyards had slight differences in soil and topoclimate. The grape–must–wine analysis also highlighted some differences between the vineyards, but the origin of some organoleptic parameters in the food chain was more evident in the comparison between V1 and V4. Terroir research showed that these two vineyards differed the most from each other from a geopedological perspective. V4 had sandy–gravelly dolomitic soil, while V1 had a finer-textured siliceous soil richer in iron oxides and clay minerals. Clay soils tend to be rich in nitrogen, which benefits wines in several ways. Grapevines use nitrogen to build essential compounds, including proteins, enzymes, amino acids, nucleic acids and pigments [72]. Furthermore, these features are related to other specific chemical and physical soil factors that are major drivers of terroir expression, influencing vine physiology (i.e., soil and subsoil temperature, pH range, soil fertility and soil water and oxygen content). Specifically, the mineral component in the phyllosilicates, mainly from metamorphic parent rocks, identifies soils with a finer texture and slightly lower pH levels. Clay minerals also improve soil fertility and CEC; therefore, they affect water retention capacity and soil temperature in a complex manner. The presence of calcite significantly increases the pH range and active carbonate soil concentration. Dolomite regulates the soil’s pH range and, if present in prevalent amounts, is conducive to coarser-textured soils and better internal drainage and aeration. However, it lowers soil fertility (CEC < 10 meq/100 g).

While other ecological factors (air temperature, rainfall and radiation) were measured as constants in the geographical model analysed, this study attributed specific ecophysiological importance to geopedological factors. Soils are the most variable element in the local terroir that can affect the typical varietal properties of the Gewürztraminer wines analysed.

4.5. Aromas in Grapes and Wines

The in-depth analysis of the individual vineyards identified micro-abiotic environmental variables, which could be the origin of the differences measured in varietal aromas. An oenological–chemical analysis found slightly different aromatic profiles in all wines. As previously explained, the terroir differences were more evident when comparing V1 and V4, which had the most varying and extreme soil characteristics. Correspondingly, the oenological products of V1 and V4 contained some key elements that differed significantly: V1 grapes had much higher amino acid concentrations (L-alanine, L-arginine, L-glutamine, L-asparagine, L-serine and L-threonine), and V1 wine had more equilibrated rose oxides and flavanol phenolic compounds with a spicy note (4-vinylphenol and 4-vinyl guaiacol) than the other wines. However, in V4′s wine, the thiol compounds with a tropical scent (3-mercaptohexanol) were significantly lower. Regarding the higher amino acid content in V1 grapes, note that these organic molecules are actively involved in winemaking processes, optimising and promoting the biochemical transformation of secondary aromas into tertiary ones [73], and could improve a wine’s aromatic potential [74]. This result is linked to the well-established evidence of aromatic typicality consistently measured over the past few decades in wines from this specific vineyard. Specifically, V1′s wine has distinguished itself over the years through a distinct and original property of evolving tertiary aromas over the course of its life.

This study on the terroir effect on wine typicity mainly focused on the various abiotic factors in the seven vineyards. The results from the targeted chemical and oenological analyses conducted on the varietal aromas of grapes and wines seem to match the differences measured in the terroir. Nevertheless, we are aware that there may be complex metabolic and biosynthesis processes in the evolving tertiary aroma compounds in wines that require future oenological studies and specialised analyses.

4.6. Geographical Traceability of Wine

The results of the present study concerned some elemental minerals measured in soils, grapes, musts and wines, indicating the presence of direct links in the entire food chain: Rb/Sr isotopic ratios and the Ba, Rb, Mn and Be contents. The study showed that the mineral composition pattern was transferred through the soil–wine system and that the differences observed in the soils were reflected in the grapes, musts and wines, but not in all the elements. These results confirm the correctness of the adopted research methods and the mineral fingerprint approach’s geographical origin traceability of oenological foods. Furthermore, the analyses indicated that many of the soil’s minerals were actively involved in winemaking processes, interacting in different ways with yeast or biochemical processes, thereby altering the minerals’ final content in the seven wines. The results suggest that in complex geo-pedological conditions, both vineyards and winemaking are important, as mixing grapes from several vineyards, even if these are quite close to each other, can already change the wine’s elemental composition.

5. Conclusions

One of the greatest research and development aspirations is to improve objective measures for terroir’s impact on wine quality. Recent literature confirms that advances can be achieved by deconstructing measurable soil and climate parameters’ effects on grape and wine aroma compounds. In this study, this incidence was assessed through analysis of both the main abiotic factors of terroirs and the Gewürztraminer cultivar’s varietal typical aroma compound. The study found that even in very complex geographical environments, a comparative qualitative-quantitative analysis of all ecological indicators makes it possible to identify each vineyard’s most characteristic abiotic factor. The results indicate that in the small research area, climatic parameters, e.g., air temperature, rainfall and solar radiation, are fairly constant among vineyards, while geopedological characteristics are more variable. Soil is here a major abiotic driver of terroir expression that affects wine quality by influencing vine development and grape ripening through soil temperature, water and mineral supply.

In Tramin, wide-ranging geopedological analyses have been extremely useful in detecting the two most different geo-mineral and physical vineyards soil conditions: Vineyard 1 has fine-textured soils of mixed mineralogy, predominantly silicate and clay minerals of glacial origin, and Vineyard 4 has mainly sandy gravel dolomite soils, predominantly carbonate minerals of local ancient debris flow sediments. These two different soils are apparent in the oenological components, flavours and aromas of their musts and finished wines: targeted oenological chemical analysis of Vineyard 1 showed higher amino acids contents and lower levels of β-phenylethyl alcohol, total polyphenols, phenolic compounds with a spicy note 4-vinylphenol and 4-vinyl guaiacol; vineyard 4 analysis revealed lower content of 3-mercaptohexanol, a thiol compounds with a tropical scent, and higher levels of free-form geraniol, 4-vinyl guaiacol and citronellol.

These results reinforce that the composition of “Gewürztraminer” musts and wines is strongly determined by vineyard site, even in a small geographic area with high variability of the terroir. This evidence also is confirmed by the winemakers themselves, who over the last decades have noted a certain delay/slowing of vegetative activity of vines growing in some finer siliceous soils and different quality and style expressions among their wines. The preliminary results obtained are promising; further confirmation would require long-term air temperature measurements for every vineyard and a larger sample of wines, preferably from several vintages. The method employed could represent an affordable terroir analytical tool in complex ecological conditions. By combining new Gewürztraminer terroir information on soil mineralogy and texture, associated with geopedological knowledge of the local soil and subsoil history, it may be possible to help local winegrowers and winemakers monitor and predict their vineyards and wines’ quality predisposition.