1. Geographical settings

The focus area of this research was the province of South Tyrol, a wine region in northern Italy (latitude 46° 30’ N) within the central-eastern part of the vast and articulated geographical environment of the Alps, which extend for more than 1,000 km from France (longitude 5° 40' E) to Austria (longitude 16° 20' E). The Alpine morphology is based on strong vegetational and microclimatic contrasts related to the almost exclusively mountainous area, with marked differences in elevation between the valley bottom and the top of the mountains, ranging from 200 m to over 4,000 m above sea level (ASL). The valleys are flanked by steep slopes and cliffs, often marked by high rocky walls, with quite large alluvial bottoms, around which the majority of the vineyards grow. There are extensive plateaus above the main valleys in the mid-mountain area. Between the valley floors and the plateaus, smaller and discontinuous morphological terraces are arranged on several levels of the valley sides. Because of the very favourable climate here, especially in the foothills, vines have been grown for many centuries at altitudes of between about 200 m and 1,300 m ASL. The studied wine territory fits into this varied geographical context and is located in the Bassa Atesina sub-region, south of Bolzano city, near the villages of Tramin, Magreid and Montan (Figure 1).

Figure 1. South Tyrol and geographic sub-regions.

Save View full size Expand inline Collapse inline

Bassa Atesina - Unterland is located in the southernmost part of the region.

2. Geopedology

The geological history of the Alpine region has defined all the geographical elements that characterise the landscapes and thus the many natural ecological factors that control the phenological behaviour of the vines. During the formation of the region’s mountains and deep valleys, erosion of the more friable rocks has made it easier to form morphological terraces, which are today cultivated with vineyards. Glaciers, rivers and landslides shaped these morphological terraces and thin layers of sediment were deposited both at the margin of the valley floor and along the steep valley sides. Here we find a wide range of Quaternary sediments, which were deposited side by side or on top of each other in the last few millennia: different glacial, debris flow, landslide, alluvial fans or alluvium sediments, as well as slope debris, colluvial deposits and eluvium often reworked by anthropogenic action. Thus, the properties of the soils have been influenced by the geological history of the area, which has shaped their physical structure, mineral composition and acidity. Dolomite, limestone and ancient sedimentary rocks mixed with porphyry and metamorphic deposits from the north of the region are present in the southernmost part of South Tyrol. Near the villages of Tramin, there are fine soils of fluvial-glacial origin from ancient sedimentary rocks, with a significant silty-clayey component coloured by iron oxides, like hematite. Calcium carbonate and dolomite minerals result in soils with a pH ranging from slightly alkaline to strong alkaline. For the restricted research area at Tramin (Figure 2), an analysis of seven vineyards with different geopedological conditions was carried out applying the Vineyard Geological Identity (VGI) method (Ferretti, 2019a). The VGI method entails the use data and information obtained from the research technologies of applied geology and the documentation of the physical, mineral and chemical characteristics of soil sediments in a comparable manner. This method is useful for discriminating and predicting the geographical components strongly connected to the ecological diversity of each vineyard.

Figure 2. Location of all the Gewürztraminer vineyards at Tramin and their SRI index (see §6. Vineyard Solar Radiation Identity (SRI) index).

Save View full size Expand inline Collapse inline

Blue markers show the seven vineyards studied using the VGI method. Red markers indicate the three environmental monitoring stations described in §4. Local climate

Ferretti (2019a) and Table 1 list the geopedological classification components of the vineyard soil sediments at Tramin, which highlight their lack of homogeneity due to the complex and varied geological evolution of this portion of the Alpine territory.

3. Regional climate

The climatic conditions of South Tyrol are closely linked to its geographically distinct location within the Alps. The size of the area and the fact that it is completely surrounded by high peaks and mountain ranges mean that it has its own protected macroclimatic conditions. South Tyrol is among the driest areas of the Alps with an annual rainfall of 600 to 1,100 mm and an unusually sunny climate, with more than 300 days of sunshine a year (EURAC Research, 2018). An interesting fact about the local climatic conditions of consequence for viticulture is the variation in air temperature with elevation, with variations of 0.6 to 1.0 °C per 100 m in the summer season (Ferretti, 2020a).

Also of consequence for viticulture could be the future ecological impact of climate change (van Leeuwen and Darriet, 2016). Climate change has been apparent in the Italian Alps and in South Tyrol in recent decades with increases in air temperatures, especially minimum temperatures (Acquaotta et al., 2014; Zebisch et al., 2018). In the last 25 years, we have measured a regular and constant temperature increase of up to 4 °C during the budding season; this induces an earlier vegetative cycle and earlier ripening and causes the harvest to take place some three to five weeks earlier (Figure 3). Several regional climate models have predicted the overall effects of individual or combined climate change-related variables (Stock et al., 2005; White et al., 2006; Hall and Jones, 2009); these experimental data are in line with the forecasts of the temperature-based model of Lebon (2002). Even the quality analyses on red grape must, which have been carried out on test vineyards in the second week of September for the last thirty years, show a regular trend: sugar content has increased from about 15 to 20 °Babo and contemporary titratable acidity has decreased from about 14 to 7 g/l (Haas, 2019).

Figure 3. Harvest day in the last 25 years of Sauvignon blanc ‘Voglar’ in the most southern part of South Tyrol.

Save View full size Expand inline Collapse inline

In the early 1990’s, the harvest normally took place in early October; in recent years, harvesting has taken place at the end of August.

4. Local climate

Because a fundamental ecological factor is the specific temperature requirements of a given grapevine variety (Parker et al., 2013), one of the aims of the research was to identify any influence of the microclimate on the ideal phenological ripening process of the grapes. The timing of phenology (budburst, flowering and veraison) and consequently the harvesting period, is driven mainly by vineyard environmental conditions, air temperature (Parker et al., 2011), rainfall and soil water content (Tesic et al., 2002), sunlight (Spayd et al., 2002) and soil temperature (van Leeuwen et al., 2018).

To cover the vast geographical area studied, the data from twelve different weather monitoring stations in the Bolzano Province in the southern part of South Tyrol were initially evaluated. However, these are located outside the vineyards and their data was therefore not found to be significant for a highly detailed study of an area in a mountainous environment.

To ensure maximum reliability in terms of microclimatic data, the more detailed part of the study focused on a single grape variety and a very restricted production area of about six square kilometres around Tramin; thus the microclimatic variables were reduced, while a statistically significant amount of vineyard data was kept. Three complete environmental monitoring stations were installed near the village of Tramin, in order to monitor and sample many environmental parameters throughout the 2019 growing season (Table 2). The three stations are located approximately 2 km apart and cover a strip of land around Tramin that stretches from north to south (Figure 2). The parameters measured by the three stations were: air temperature, solar radiation, dew point (steam to liquid), vapour pressure (liquid to vapour), relative humidity, rainfall, foliar wetting and wind speed and direction, as well as soil and subsoil salinity, temperature and humidity at various depths.

5. Soil temperature and moisture

A further environmental stress analysed in the study was soil temperature, as it influences the biochemical and nutritional processes of the plant. The heat absorbed by the first layers of soil through solar radiation is the energy charge that contributes to biochemical reactions of plants. Soil temperature in the root zone impacts phenology and timing of ripeness (van Leeuwen et al., 2018); it depends on water content and energy balance (which is related to soil mineralogy, colour, coverage and solar radiation on the soil), as well as to the vineyard's altitude and topography. Warm soil in the root zone generally results in better wines (Bodin and Morlat, 2006; van Leeuwen, 2001; Tesic et al., 2002). Temperature fluctuations are induced by a number of factors, such as a convective exchange with air, irradiation and rainwater. Ground temperature was measured continuously at the foot of each monitoring station by 12 sensors placed at 10 cm intervals down to a depth of 1.2 m. Humidity sensors were installed in combination with temperature sensors. Soil moisture sensors determine volumetric water content (VWC) by measuring the dielectric constant of the soil using capacitance technology and soil temperature.

6. Vineyard Solar Radiation Identity (SRI) index

The South Tyrol wine-growing region is characterised by numerous small production sub-zones, with more than 26,000 vineyards which differ greatly in their geographical features, influencing the quality of the grapes. Geographical features (altitude, exposure, inclination and micro-climatic conditions) can mainly be identified in the topoclimate and as a whole they constitute an important natural ecological resource of a given area. In a recent study, the new solar radiation identity (SRI) method (Ferretti, 2020a) was proposed to accurately classify the topoclimate in complex geographical environments. The aim of the SRI method is to make it easier to evaluate the impact of the vineyard’s geographical conditions on both the vegetative growth of different grape varieties and the quality of the wine they produce.

The SRI index is a new approach to applied research and decision making in the viticultural sector; it was implemented to classify the geographical identity of vineyards in a more precise and comparable way. It is derived from a scientific analysis of vineyards and it is used for the discrimination and synthesis of complex environmental conditions of vineyards in mountainous areas. The topoclimate consists of numerous independent variables related to the vineyard, namely: slope, orientation, position and exposure, mountain shade and the position and zenith of the sun on the horizon. SRI goes beyond just an irradiance rate by giving a specific vineyard an identity value related to the topographic shape of the vineyard (derived from a high detail digital terrain model (DTM)) and the time variable. Thus, every index accurately takes into account both the changing height of the sun and the shading scenarios. SRI’s database is also accurately related to the real size of every mono-varietal vineyard, which in our case represents the reference lot for the production of a homogeneous batch of wine.

With its highly detailed geospatial analysis, the SRI method, represents the whole viticultural area and catalogues the topoclimatic value of each single vineyard within a simple numerical index rate (Figure 4). All the regional vineyards were classified within the two extreme values of the SRI index and normalised between 0 and 100, according to the following standardisation algorithm:

SRI = ((SR_v - SR_min) / (SR_MAX - SR_min)) × 100.

SR_v = potential solar radiation received by the single vineyard measured in kWh/m² for the grape growing period of about 120 days (D'Onofrio, 2011), assumed to be between 1st May and 1st October. SR_min and SR_MAX are respectively the lowest and the highest values within the wine region. The calculations were executed with a raster cell resolution of 5 × 5 m, precisely shaped on each vineyard’s real topographical position and then averaged to give a representative value of its topoclimatic identity. It is thus possible to obtain a complete database for each vineyard, including those of the Gewürztraminers studied here, with which to analyse, verify and precisely compare its ecological conditions. These quantitative data help to characterise the potential of each vineyard and are useful for discerning the influence of ecological factors on the quality of musts and wines. The index rate can thus be linked to the biosynthetic activity of the vines, as well as to the quality of the vineyard’s grapes and wine.

Figure 4. Comparative image from almost the same point of view between vineyards of the Söll geographical unit and their SRI indexes.

Save View full size Expand inline Collapse inline

This way of representing them simplifies the distinction between the different geographical predispositions of the vineyards.

7. Grape must sugar content

To ensure the reliability of the study, the territorial research focused on a single grape variety and analysed a total of 315 data of Gewürztraminer grape musts, harvested in 256 vineyards located in the Bassa Atesina sub-region (Figure 5), from a total area of 69.2 hectares, which represents 12 % of the total Gewürztraminer cultivated in South Tyrol. Grapes were harvested at altitudes of between 208 and 612 m in which the vineyards’ SRI index varies from 47 to 84. In 2019, the grape harvest took place from 20 September to 15 November during the weeks that were dedicated to late harvests.

The ripeness of the grapes was determined by analysing the sugar content of the must. Assuming that most of the solids are sucrose, must weight can be converted directly into amount of dissolved sugar, which is close to the true value and defines a first wine quality classification (Peth et al., 2017). Many countries use different measurement units of the must's weight (Hopfer et al., 2015). In Austria and South Tyrol, sugar content in must is measured in percent by weight (g/100 g) with the Klosterneuburger Mostwaage. Their scale shows Klosterneuburger sugar degrees (°KMW), also adopted from Italian viticulture as Babo degrees.

Must analysis was carried out immediately after the delivery of the grapes to the cellar, on must coming from the first pressing. Sugar content was measured on production batches of grapes. Such data are used by wineries to define the quality and economic value of the lot. This testing procedure was chosen to verify the practical applicability of the research methods to general practice in wineries and real wine production chains located in a large area of vineyards.

A more detailed analysis of must quality was carried out on a restricted area near Tramin - on vineyards from the geographical sub-zone of Söll, St. Jakob, Plon and Pinzon. A dataset of 48 selected vineyards from a single winery was analysed. The cultivation practices in the vineyards, although not exactly the same, are highly comparable. The harvest was carried out on 3 and 4 October 2019. These dates, carefully chosen by the oenologist, were when the desired degree of optimal ripeness of the grapes had been reached. The vineyards lie at altitudes of between 232 and 535 m, have a SRI of between 64 and 80, comprise a total area of 12.5 ha and provided a total of 87 t of grapes in our study.

Figure 5. Base map of the new geographical units proposed for Bassa Atesina and the most southern part of South Tyrol, listed in their respective municipalities.

Save View full size Expand inline Collapse inline

Blue box = the location of the Gewürztraminer vineyards harvested in 2019 between 20 September and 15 November; red box = restricted area near Tramin with vineyards harvested on 3 October and 4 October (see Figure 2)

1. Air temperature

The harvesting period is mainly driven by growing season air temperature (Parker et al., 2011) and the specific temperature requirements of the grapevine variety (Parker et al., 2013). The long-term analyses for all three environmental monitoring stations installed at Tramin, indicate that their seasonal temperature trends correspond; daily averages increased between April and May and the peaks in temperature occurred between the end of June and mid-August. The data comparison between Plon and Söll confirms a reduction in air temperature with altitude of approximately 0.9 °C per 100 m (Table 3). On the one hand, this matches the typical regional values during the summer period, which show temperature decreases of 0.6 to 1.0 °C for every 100 m in altitude. On the other hand, the research shows that neither the altitude nor the relative reduction in temperature significantly affected the day of harvest or the quality of the must samples from the Bassa Atesina. In fact, only a weak positive relationship (Pearson's r = 0.21, p < 0.05) was confirmed between the vineyards' altitude and the sugar content of their musts (Figure 6). We found that the harvest date is conditioned more by both rainy days and the schedule of the winery's activities. The harvest of vineyards at altitudes of between 210 and 600 m was sometimes carried out on the same day. Nevertheless, the winemaker confirmed that in the small area examined at Tramin, the trend in seasonal air temperature or the vineyard altitude from about 230 to 550 m affected the timing of phenology in terms of budburst, flowering and veraison, delaying it by five days at higher altitudes in comparison to grape vines on the valley floor.

Figure 6. Comparison between harvest date and vineyard altitude with a statistically significant (r = 0.21, p < 0.05) correlation. The harvest day may follow the wineries’ schedules. The vintage at different altitudes was sometimes carried out on the same days.

Save View full size Expand inline Collapse inline

The detailed analysis of the temperatures in the vegetative period highlights the considerable microclimatic temperature differences within a small area of only 6 km² around Tramin. Table 3 shows that while the Am Sand station - which is at a lower elevation than the others - has higher daily peaks in temperature, the minimum and average temperatures are abnormally low. This can be explained by comparing the day/night temperature range between the stations. At Am Sand there is a higher diurnal temperature variation (Figure 7). Here, the night-time temperatures are between 2 and 3 °C lower than in Plon. This can be attributed to the particular microclimate, typical of an Alpine mountain environment. In particular, local cold air currents sink during the night from the high mountain cliffs made of dolomite rocks which overlook the village of Tramin to the west. These fresh air currents are channeled through the narrow valley of the Höllental Torrent and reach the Am Sand station placed at the end of the valley. Daily temperature differences between the three monitoring stations reach up to approximately ΔT 6 °C, with differences between the maximum temperatures of up to +1.5 °C and between the minimums down to -3.5 °C.

Figure 7. Diurnal temperature variation during the 2019 pre-harvest month.

Save View full size Expand inline Collapse inline

The graph shows differences between the environmental monitoring stations. The Am Sand station is particularly noteworthy, recording the highest diurnal variations between low and high temperatures.

2. Rainfall

There were 78 rainy events in the 197 days monitored between 23 March and 31 October 2019. The rain events registered by the three monitoring stations were simultaneous and had the same intensity. For some sporadic events, different amounts of rainfall between the stations were recorded, due to the mountainous area, which favours local thunderstorms with of differing strengths. The total quantity of rainfall at each of the three stations in the April to October period was very similar: Söll = 686 mm, Plon = 683 mm and Am Sand = 664 mm.

3. Solar radiation

Solar radiation was measured using the two weather stations in Söll and Plon. Solar data was sampled continually on-site with a pyranometer from May to October 2019.

Solar irradiance is measured in W/sqm and represents the power of the sun’s rays falling on a given surface in the form of electromagnetic radiation. Its trend over the whole measured period was highly cyclical between the night-time minimum and the daytime maximum (Figure 8). There was a seasonal trend for an increase in spring/summer and a decrease in autumn/winter. The Söll station, located in the northern part of the area, recorded a higher level of irradiance than the Plon station, which was located in the south.

Figure 8. Solar radiation (SR) graphs: left the daily mean values measured in Plon and Söll between August and October 2019; on the right, values from 30 September. The total SR measured in the afternoon was +13.5 % at Söll (2,720 W/sqm) compared to Plon (2,350 W/sqm). The comparison indicates higher energy at Söll.

Save View full size Expand inline Collapse inline

On sunny days a maximum difference in energy W/sqm of up to 14 % was measured in favour of Söll. These differences are due to the dissimilar topoclimate situations. The geographical evaluation of the two vineyards indicates that both are oriented towards east-south-east, but the first is less steep and covered daily by late-afternoon shade from the mountains.

4. Soil moisture

The change in soil moisture measured at the Plon and Söll stations precisely follows the cyclical nature of the rain events. The humidity level varies seasonally: in the subsoil it is around 18 % in late summer, increasing with autumn rains and frost and at its highest in winter, settling at around 30 %. Early harvest dates likely correspond to the highest levels of water shortages at the end of the season. Throughout the growing season the soil in Plon is constantly 8 % wetter than that in Söll. This is in line with the lower permeability of Plon’s silt loam soils compared to Söll’s loam soils. It also explains the lower soil temperature and higher thermal capacity of Plon described in the following sub-section, since more energy is needed to heat the volume of soil richer in water.

5. Soil temperature

In local wine-growing practices, a permanent grassy cover is left on the soil, which is never ploughed or directly exposed to air. The temperature data of the most superficial soil show a diurnal trend similar to that of the air temperature, with minimums during the night and maximums in the afternoon. The soils heat up in the morning and respond to the solar radiation they receive. Also, the rains have an instantaneous impact on the soil temperature and quickly cool them, even at depth. The data collected show that the daily temperature range in the more superficial soil depends mainly on solar radiation, as well as on the physical and mineral nature of the soil, in particular texture and moisture. With the same energy availability, a wet soil heats up less than a dry one. The air temperature, however, acts on the soil with a monthly and seasonal return of time. It can be observed that the air temperature in Söll is on average about 1 °C lower than in Plon, but the soil at 10 cm depth in Söll is on average 1 °C warmer than in Plon (Table 3). Thus, the Söll vineyard has warmer soil, although it is approximately 100 m higher than Plon and the air temperature is colder. The temperature of the soil depends not only on the solar radiation that reaches it, but also on its physical properties of texture and humidity. As described above, Plon’s soil is finer and wetter than that in Söll.

Figure 9 illustrates the thermal soil model at Söll and Plon. First of all, the direct relationship between solar radiation and temperature can be observed and the two soils also have different thermal diffusivity. In the morning, Plon still tends to cool down, despite the increase in solar radiation energy, while Söll responds to it more directly and increases its temperature as soon as the sun’s rays reach the vineyard. The soil in Söll is warmer on the surface than in Plon. Söll has higher specific heat and accumulates more energy, which then returns to the plants and the environment.

The subsoil that regulates the temperature in the root zone at a depth of below 40 cm, stays at a constant temperature in the short term, without daily fluctuations. The surface soil layer acts as an insulator, stopping the direct influence of solar radiation at depth. In the long term during the year, the thermal trend is divided into three main phases: heating (April to June), maximum temperature (June to August) and cooling (September to December). The thermal behaviour of each of the two stations is seasonally similar, but they differ in absolute temperature values in the medium- or short-term. Söll has a faster response to seasonal air temperature than Plon: the subsoil heats up more and faster in summer and cools down more in winter.

Research has found that the temperature in the subsoil of Söll is always lower than in Plon, with differences of up to - 1.5 °C. By contrast, near the surface the soil of Söll is warmer than that of Plon - on average 1 °C during the whole growing period of the vine.

Figure 9. Comparison of the thermal model in Plon (red lines) and Söll (blue lines), average temperature per hour between 15 April and 15 October 2019.

Save View full size Expand inline Collapse inline

Lines and arrows represent the daily temperature cycle in the soils at a depth of 10 cm and in the subsoils at -110 cm. Close to the surface the soil is warmer and the temperature trend is cyclical, linked to solar radiation. The Söll soil is warmer and responds more directly to solar radiation. By contrast, deep down there is no influence of solar radiation on the temperature and the subsoil in Söll is colder.

6. Grape must quality

The quality of the grapes is evaluated by examining the sugar content of the must on harvest day. The results of the analysis of all these data indicate quite a strong correlation between harvest date and sugar content, which is more evident in the minimum values. An increase in sugar over time can be seen with a corresponding progressive physiological ripening of the grapes (Figure 10). The late harvest increase in sugar is instead related to the concentration of sugar resulting from the dehydration of the fruit (Guillaumies et al., 2011), which is superimposed on that from the physiological ripening process of the grapes. The harvest data trend, in particular the physiological maturation highlighted in Figure 10 by the sugar rise over time confirms the reliability of the data collected and the correct direction of our research. The distribution of the total data cloud is still very dispersed; however, a medium positive relationship emerged between harvest date and sugar content with a linear regression from the estimated model, y = 0.274x – 1,175.8, Pearson's r = 0.33 and p < 0.001. This is consistent with the local geographical situation and the large sample of data analysed; considering that plant responses to abiotic stresses are dynamic and complex (Cramer et al., 2011), it is thus not fully linear. There are important geographical differences in the Alpine study area and, above all, the wine-growing and wine-producing choices of the wineries had a fundamental influence on the quality of the harvested products; for example, grape picking did not always take place at the moment of the optimal ripening stage of the grapes for each vineyard; the harvest was sometimes interrupted by rainy days, which results in erratic grape ripening, or was dependent on winepress availability and cellar scheduling (sometimes the delivery of the grapes is scheduled within a narrow time slot, during which the winepress can only be used for one grape variety at a time).

Figure 10. Sampled must sugar content and harvest date.

Save View full size Expand inline Collapse inline

The red line marks the physiological ripening limit of the Gewürztraminer variety for 2019 in Bassa Atesina. Late harvests of the last weeks can be distinguished in November. The blue shaded area highlights the sugar content and ripening differences in the sub-region; these are related to the many ecological factors and viticultural practices of this wine zone. Bars on the x axis represent rainfall.

In order to filter the analysed data from the afore-mentioned anthropic and environmental variables, or from cryptogamic diseases and to better finalise the research on ecological indicators, a smaller and more controlled area around the village of Tramin was studied. We analysed 48 vineyards which each have a single wine cellar. They apply uniform and more controlled viticultural procedures and have an espalier planting structure, with all grapevine rows oriented approximately north-south. The quantity of grapes produced in each vineyard is about 0.7 t/ha. The date of harvest was chosen carefully by the oenologist to occur on 3 and 4 October 2019 when the desired optimal ripeness of the fruit had been reached. The results of the analysis revealed a clear correlation between sugar content and SRI topoclimate index for this more restricted, homogeneous and controlled geographical context (Figure 11). The graphical correlation was evaluated using a multivariate linear statistical analysis, which best reflects our study case, as it has more than one dependent variable. A strong statistical relationship between the two variables SRI and sugar is expressed by the linear regression with estimated model y = 0.0805x + 15.145, Pearson’s r = 0.69 and p < 0.001.

Figure 11. Sugar ripeness and SRI index for the vineyards at Tramin harvested on 3 and 4 October 2019.

Save View full size Expand inline Collapse inline

The graph shows a range of data strictly regulated by ecological factors. In particular, we observe an increase in sugar content with an increase in SRI, which represents the topoclimatic identity of each vineyard.

1. Air temperature

The results of the microclimatic analysis show a significant diurnal air temperature variation within the small area studied at Tramin. The air temperature data collected in this research can be used as a qualitative ecological indicator of the area, but these cannot be quantitatively traced back to all the vineyards. However, quite variable microclimatic conditions were found, which could affect the phenological response of the local grapevines and, therefore, define a different maturation rate and quality of the musts analysed in the research. Thus, for the study area, some correlation between air temperature and sugar content of the grapes at harvest may have been possible and therefore have been checked. In view of this, SRI and grape sugar content were verified, specifically in a few vineyards in the Höllental valley, but no evidence of such effects on must sugar content was found. However, only three of the vineyards studied are along the Höllental, thus the data population was too small for a statistical test. It should also be noted that Gewürztraminer is not grown on the valley floor, where higher temperature fluctuations are recorded, as this area is not traditionally suitable for its growth. However, it is possible that the above-mentioned microclimatic conditions could affect the organoleptic features of grapes and wine, which have not been studied in our research.

2. Rainfall

The cumulative amounts of daily rainfall are very similar within the restricted area around Tramin. Even in the months before harvesting, the rainfall amounts were homogeneous, (Figure 10). Therefore, differences in rainy days or amounts of rainfall could not be linked to differences in ripeness within the observed restricted wine area. However, it should be noted that when comparing vineyards in which harvesting took place on different days, rainfall may still represent a selective ecological element for vine biosynthetic activity; vineyards can therefore be differently affected by sunny or rainy weather conditions near the time of harvesting. In fact, rainy, overcast or low-temperature weather conditions can affect physiological grape ripening and result in similar sugar content of musts harvested on different days before or after the rain. We can conclude that rain does not represent an ecological variable for the qualitative analysis of musts pressed on the same day, but it does represent a variable environmental factor for grapes harvested in locations distant from each other and on different days. However, as described below, in this kind of data analysis it is important to always take into account the fact that soils have differing capacities for retaining the same amount of rainfall.

3. Radiative energy

The fact that Söll receives more solar energy than Plon seems to be in contrast with the average air temperatures, which are approximately 0.8 °C lower than those in Plon. However, the sampled environmental data show that the solar radiative energy reaching the vineyard and air temperature are two separate ecological indicators that can differ significantly in a mountainous environment. A further important result is that the SR value difference of 14 % registered at the weather stations is fully in line with the SRI index value calculated using the Solar Radiation Identity method (Ferretti, 2020a). The SRI index for Söll is in fact 15 % higher than that for Plon. The instrumental measurement of the SR value confirms the correctness and reliability of the SRI index method. This index can be efficiently used as an indicator of both the environmental stress to which the vines are subjected and the ecological potential of each vineyard. We can use the SRI index as a practical and suitable reference to predict the biosynthetic activity of the vines. Because solar radiation interacts with several variables (e.g., grape variety, soil and its type of cover, training system and row orientation, etc.), the SRI index should be contextualised to the experimental conditions to obtain more accurate information.

4. Soils humidity and temperature

In our study, the soils of Tramin were found to be physically affected by their humidity and temperature. Soil moisture information can be used qualitatively in our study and be linked to the extremely variable characteristics of the local soils, but because we only have a few measuring stations, such information cannot be quantitatively traced back to all the vineyards observed.

The differences in soil temperature are related to the physical characteristics of the soil, in particular on the texture, which is the cause of lower permeability but higher moisture in Söll soil compared to that of Plon. Montheith and Unsworth (1990) confirm that thermal conductivity is higher in porous soil, compared to a clayey soil containing the same amount of water. The minerals in the soils of Plon and Söll are similar and have less influence on the differing thermal behaviour of the studied soils. The analysis of soil temperature revealed it to be of ecological importance in the local soils, which can vary significantly in the vineyards around Tramin. Research has also shown that soil temperature is causally related to the amount of solar radiation received by the vineyard. Our results show that soil temperature predictions in the short-term were least sensitive to air temperature changes occurring during the day or week. Therefore, soil temperature can also be better qualitatively described by the SRI index, in order to conduct a predictive evaluation on the vineyards’ biosynthetic processes and their topoclimatic potential.

5. Sugar content

Figure 11 shows that sugar content maintains a variation range of approximately 1° KMW at the same SRI. This is an admissible inaccuracy for the research carried out due to the complex, natural, physiological processes being dependent on the environmental diversity that differentiates the vineyards around Tramin. Some of these inconstant ecological factors were qualitatively identified in our study: the geopedological differences in the soils, microclimate and air temperature, soil temperature, exposure of the bunches after thinning and, last but not least, the method of sampling the must. Regarding the latter, it should be noted that we collected sugar-ripeness data on large lots of grapes in order to verify the practical applicability of the research to a wide geographical area and to real winemaking procedures. Future research refinements could include considering more controlled situations and testing the SRI method in detail in other geographical contexts and on other wine varieties. We also believe that in order to be more technically useful the analysis of ecological indicators must include the evaluation of practical situations on an industrial scale and not only in micro-vinification conditions.

6. Vineyard height

A further interesting result of the research, which should be examined in future studies on Gewürztraminer and other wine varieties, is the fact that only a weak correlation was found between vineyard altitude and the sugar content of the grapes. It is therefore possible that some of the direct correlations between altitude and sugar content reported in other studies (Pedri, 2014) dealing with vineyards scattered over large areas depend more on other ecological factors which affect physiological maturation, such as solar irradiation, local air temperature, microclimate and soil type. In fact, the topoclimatic analysis of all our vineyards indicates a medium relationship between SRI and height, expressed by an estimated model y = 64.15 + 0.027x, Pearson's r = 0.4 and p < 0.001. The results of recent regional level SRI analyses (Ferretti, 2020a), have shown that solar irradiation on vineyards in the South Tyrolean wine region tends to increase with altitude and this ecological evidence might better explain some altitude-related physiological plant responses.

7. Vineyard topoclimate

The present study has found that the variability in sugar content is mainly regulated by a single ecological factor: the vineyard’s topoclimate. We have also tested that the topoclimate is adequately represented by the SRI index. The accuracy of the SRI index allows incidence on ripening to be quantitatively estimated: an increase in SRI of 10 points corresponds to an increase in sugar content of about 0.8 °KMW. As the ability to maintain high quality wine and to adapt to climate change will depend on our aptitude in predicting the significance and influence of ecological indicators, a prediction model based on SRI indexes is proposed.

The strong correlation between SRI and sugar having been confirmed by the study, it is now possible to develop a prediction model for SRI/sugar content, even outside the experimental data range of this research: the tested vineyards by Tramin have SRI indices of between 64.8 and 82.6, while the SRI values of the 2,732 Gewürztraminer vineyards in the whole of the South Tyrol wine-growing region vary between 40 and 92.

The proposed prediction model is based on the analysis of finite subfamilies of variables obtained within the measured datasets, with a particular focus on quantitative variables for ranges close to SRI = 3.5. Box plot-type statistical graphs in Figure 12 show more precisely the shape of the distributions of these data subfamilies, representing the statistical data listed in Table 5.

Figure 12. Analysis of finite sub-families of data, combined chart of box plots (dark green) and density pots (light green), with connecting lines between groups of means (dots) and medians (continuous). The dot corresponds to a statistical outlier.

Save View full size Expand inline Collapse inline

A correlation analysis by linear regression of the values thus obtained for each sub-family of data allows us to define the central value of the distribution, with the confidence levels indicated in Table 5 and the following linear correlation of the mean values:

Sugar content = 15.675 + 0.0707 SRI; Pearson’s r = 0.91; p < 0.5; σ = 0.46

To complete the model, the estimated parameter must be accompanied by a range of plausible values based on the variability σ of the data populations, with a 95 % confidence level as shown in Figure 13.

This model aims to estimate the physiological response of a vine variety - in this case the Gewürztraminer - in a wider topoclimate stress condition, which is typical of the mountainous South Tyrol winegrowing region.

Figure 13. Prediction model of the Gewürztraminer grape sugar ripeness based on an SRI curve analysis.

Save View full size Expand inline Collapse inline

Black line shows the mean trend of expected values, red lines define upper and lower limits of statistically significant data (equal to 95 % of normal distribution). Dots represent the measured values. The blue band between red lines identifies a deviation range due to the variability of the natural physiological processes between vineyards, due, for example, to microclimate, anthropic factors (viticultural practices, clone, rootstock) and soil (texture, minerals, temperature, humidity).

The aim of the study is to integrate the monitoring and assessment of ecological and environmental indicators into management practices. This model could assist the search for new approaches to adaptive and more flexible crop management, based on the equilibrium of the local dynamic ecosystem. New strategies would also need to be implemented to find ways of adapting vines to future climate conditions in the Alpine area, according to a well-known environmental scenario. For example, a vineyard with a low SRI could facilitate the adaptation of current cultivars to an increase in temperature; in contrast, the growth of grapevines at high altitudes requires more energy and therefore vineyards with higher SRI values which are more appropriate for the cultivar should be sought. Low SRI values can also be considered to favour the acidity so sought after for biodynamic cultivation or for sparkling wines.