Introduction

Verifying the geographical origin of agricultural products is becoming increasingly important, especially due to regional certification. Inorganic chemical analysis increasingly plays a key role in these purposes. While determining all processes influencing element concentrations in a given product is often complicated, the system is usually treated as a "black-box", and multivariate statistical analysis is used. On the other hand, isotopic ratios of heavy elements have the potential to distinguish provenance that may have been masked in examining elemental concentrations alone. Strontium is particularly well suited for this type of investigation. It is sufficiently abundant in rocks and plants, and its isotopic composition can provide a fingerprint of the nutrient sources. Strontium isotopes have been proposed and/or successfully used as a source area tracer for several foods, including wine (first suggested by Horn et al., 1993, juice (Rummel et al., 2010), or milk (Crittenden et al., 2007).

Since the early study by Horn et al. (1993), numerous studies have examined the applicability of radiogenic isotope ratios (⁸⁷Sr/⁸⁶Sr) of strontium to wine origin, as described by (Boari et al., 2008; Epova et al., 2019; Moreira et al., 2017; Tescione et al., 2018; Tescione et al., 2020). The chemical and biogeochemical behaviour of Sr is similar to that of Ca; however, in contrast to stable Ca isotopes (⁴⁴Ca/⁴⁰Ca), the ⁸⁷Sr/⁸⁶Sr isotope ratio varies considerably in natural environments, and this variability is geologically (i.e., geographically) systematic. Many authors suggest that radiogenic Sr isotopes could serve as an excellent fingerprinting tool for tracing the geographical origin of wine (Almeida and Vasconcelos, 2004; Horn et al., 1993; Marchionni et al., 2013; Petrini et al., 2015; Tescione et al., 2020), but the details of the Sr isotope connection between rock, soil and wine are not straightforward.

The final ⁸⁷Sr/⁸⁶Sr composition of the wine results from isotopic mixing of several components: soil bioavailable Sr originating from bedrock, Sr dissolved in groundwater, atmospheric input (rainwater and sea aerosols), and Sr from the anthropogenic additives such as agrichemicals. The Sr isotope ratio of products that combine several regionally different raw materials with different isotope compositions would be a weighted average, considering variability in the Sr concentration of each component. Reverse identification of two or more end members of such a mixture in a one-dimensional isotope system (such as ⁸⁷Sr/⁸⁶Sr) is impossible (Albaréde, 1995). Still, the Sr isotopic composition of rainwater and seawater is known, as are the overall compositions of various rocks.

The Sr isotopic fingerprint is regionally (geologically) characteristic but not unique. Therefore, it cannot be used to identify the exact place of origin, but it can confirm (or reject) potential candidate sites and help identify potential candidate sites.

This study aimed: (1) To examine the Sr radiogenic isotopic composition of wines from South Moravia, Czech Republic and compare them with the isotopic composition of bulk disturbed soil samples, bio-available Sr, and a groundwater sample from a local well. (2) To present a simple model of ⁸⁷Sr/⁸⁶Sr isotopic composition of wine formulated from a geological point of view. (3) To assess the limitations of ⁸⁷Sr/⁸⁶Sr as a tool for determining the geographical origin of wine.

Materials and methods

1. Sample selection

Figure 1. A. Location of the studied area in the context of Central Europe. B. Wine region “Morava” (Moravia) with sub-regions (Znojmo, Mikulov, Velké Pavlovice, Slovácko (mapavina.cz/online) with studied localities. C. Geological sketch of the Moravia Region, Czech Republic (https://mapy.geology.cz/geocr50/). All vineyards lie on loess bedrocks.

Save View full size Expand inline Collapse inline

Table 1. Sr isotopic composition of wines from Southern Moravia, Czech Republic.

View popup Expand inline Collapse inline

	Winery	Sample	Locality	Year	Grape variety	Group	87Sr/86Sr	2 SE
Mikulov subregion	Ludwig	vino 25	Sedlec	2019	Chardonnay (Ch)	A	0.709165	0.000012
	Ludwig	vino 26	Sedlec	2019	André (An)	A	0.709063	0.000010
	Ludwig	vino 27	Sedlec	2019	Zweigelt (Zw)	A	0.708612	0.000008
	Ludwig	vino 28	Sedlec	2018	Welschriesling (RV)	A	0.709338	0.000010
	Ludwig	vino 11	Sedlec, Stolová hora	2019	Rheinriesling+Aurelius (Fr+Au)	A	0.709200	0.000014
	Vican	vino 2	Pod Mušlovem	2016	Blaufränkisch (Fr)	A	0.709549	0.000014
	Vican	vino 13	Pod Mušlovem	2017	Blaufränkisch (Fr)	A	0.709550	0.000012
	Vican	vino 15	Pod Mušlovem	2018	Blaufränkisch (Fr)	A	0.709493	0.000009
	Vican	vino 16	Pod Mušlovem	2019	Blaufränkisch (Fr)	A	0.709320	0.000012
	Ludwig	vino 8	Janův vrch	2018	Muscat Moravian (MM)	C	0.709324	0.000011
	Ludwig	vino 10	Stará hora	2019	Neuburger (Ng)	C	0.708392	0.000009
	Marcinčák	vino 3	Janův vrch	2017	Rheinriesling (RR)	C	0.709165	0.000014
	Marcinčák	vino 4	Janův vrch	2018	Rheinriesling (RR)	C	0.708952	0.000011
	Marcinčák	vino 5	Janův vrch	2019	Rheinriesling (RR)	C	0.708253	0.000011
	Marcinčák	vino 6	Slunečná	2016	Rheinriesling (RR)	C	0.709097	0.000012
	Marcinčák	vino 7	Slunečná	2017	Rheinriesling (RR)	C	0.709114	0.000011
	Marcinčák	vino 9	Slunečná	2018*	Rheinriesling (RR)	C	0.708443	0.000012
	Marcinčák	vino 12	Slunečná	2018*	Rheinriesling (RR) duplicate	C	0.708431	0.000012
Great Pavlovice subregion	Jedlička	vino 18	Terasy	2019	Rheinriesling (RR)	B	0.709414	0.000027
	Jedlička	vino 19	Terasy	2018	Rheinriesling (RR)	B	0.709600	0.000011
	Jedlička	vino 20	Terasy	2017	Rheinriesling (RR)	B	0.709402	0.000013
	Jedlička	vino 21	Terasy	2016	Rheinriesling (RR)	B	0.709638	0.000013
	Jedlička	vino 22	Veselí	2020	Müller-Thurgau (MT)	B	0.708988	0.000010
	Jedlička	vino 23	Veselí	2019	Müller-Thurgau (MT)	B	0.708960	0.000014
	Jedlička	vino 24	Veselí	2018	Müller-Thurgau (MT)	B	0.709352	0.000012
	Ludwig	vino 17	Kobylí	2019	Cabernet Moravia (CM)	B	0.709523	0.000011
	Ludwig	vino 14	Němčičky	2019	Solaris (Gm)	D	0.709498	0.000010
	Ludwig	vino 1	Němčičky	2020	Solaris (Gm)	D	0.708527	0.000010

2 SE: two standard error, * two different bottles of vine.

Twenty-six samples of mono-cultivar variety and one blended (vino 9 sample) commercial wines from four wineries were selected for this study (Table 1). Samples were aliquoted from bottled wine. Due to the geographical location of the vineyards, we divided wine samples into four groups according to their location: A, B, C, and D (Figure 1B,C). Wines were produced from 11 different grape varieties grown in Southern Moravia, Mikulov (group A: 9 wines and C: 8 wines), and Velké Pavlovice (group B: 8 wines and D: 2 wines) subregions (Table 1, Figure 1). Sample selection was primarily based on the ability to match vineyard location with commercially produced wine accurately; inter-variety differences were of secondary concern. Loess forms the geological substratum for all samples. The studied vineyards were founded after 1989, and all four wineries use similar methods of wine production.

Figure 2. Comparison of 87Sr/86Sr isotopes in bulk soils and their leachate. Field photographs of the studied soil samples from Sedlec u Mikulova, Southern Moravia.

Save View full size Expand inline Collapse inline

Table 2. 87Sr/86Sr composition of leachate soils and bulk soil samples from Sedlec, Mikulov sub-region.

View popup Expand inline Collapse inline

Sample	Depth, cm	Locality	Sample preparation	87Sr/86Sr ± SE
VN30	30	VN	leachate	0.708444 ± 14
VN31	30		bulk sample	0.713801 ± 14
VN60	60	VN	leachate	0.708821 ± 18
VN60	60		bulk sample	0.714684 ± 12
VN90	90	VN	leachate	0.708958 ± 14
VN90	90		bulk sample	0.715323 ± 12
VN200	200	VN	leachate	0.709460 ± 10
VN200	200		bulk sample	0.718377 ± 16
groundwater from the local well				0.709219 ± 14
open-area precipitation from Erban Kochergina et al. (2021)				0.7097–0.7098

Group A is the main focus of this study, as sampling of potential isotopic reservoirs in the local environment is the most complete for this region. In 2020, soil samples were taken from the village of Sedlec u Mikulova locality (Group A) at 30, 60, 90, and 200 cm depth intervals directly from the vineyard. A groundwater sample from a 4 m deep well approximately 1.1 km from the soil sampling point was taken for comparison with soil leachates (Table 2, Figure 2). Soil samples were not collected from the other group localities. However, due to the similar bedrock, climatic and hydrological conditions, we infer that other sites' soil conditions and geochemical characteristics are similar to those of Sedlec u Mikulova (Group A).

The shallow vine roots of Group A Rheinriesling cultivar (RR) were observed to a depth of 40 cm, and the deep roots exceeded a depth of 170 cm.

Results

We determined strontium isotopes in 27 wine samples (Table 1, Figure 3). The ⁸⁷Sr/⁸⁶Sr ratios for the groups are as follows: Group A wines 0.7086–0.7096 (mean 0.7093), Group B wines 0.7090 – 0.7093 (mean 0.7094), Group C wines 0.7083–0.7093 (mean 0.7088), Group D wines 0.7085, and 0.7094.

Figure 3. Strip boxplot (Janoušek et al., 2016) showing a compilation of 87Sr/86Sr values for selected wine samples from different groups, as defined in the Material and methods section (Figure 1B).

Save View full size Expand inline Collapse inline

We analysed 10 samples of Rheinriesling (RR) cultivar from three localities: Januv vrch (3 samples from 2017, 2018, 2019), Slunecna (3 samples from 2016, 2017, 2018) and Terasy (4 samples from 2016, 2017, 2018, 2019). These samples show differences in Sr isotopic composition according to vintage. The most heterogeneous RR samples are from Januv vrch with a maximum ⁸⁷Sr/⁸⁶Sr ~ 0.7092 in 2017, ⁸⁷Sr/⁸⁶Sr ~ 0.7090 in 2018, and a minimum ⁸⁷Sr/⁸⁶Sr ~ 0.7083 in 2019; the relative standard deviation (RSD) is 549 ppm. A similar trend, where older samples have more radiogenic composition, was observed in samples from Slunecna. Samples from 2016 and 2017 are almost identical, with ⁸⁷Sr/⁸⁶Sr ~ 0.7091, and a sample from 2018 has ⁸⁷Sr/⁸⁶Sr ~ 0.7084. The RSD for this locality among sampled vintages is 441 ppm. Samples of RR from Terasy are less heterogeneous (0.7094–0.7096), the differences in ⁸⁷Sr/⁸⁶Sr are in the fourth decimal place, and the RSD for four samples is 150 ppm.

A sample of the Blaufränkisch (Fr) cultivar from 2019 from Pod Muslovem also shows less radiogenic Sr isotopic composition (~ 0.7093) than older samples from 2016 to 2018, with ⁸⁷Sr/⁸⁶Sr ~ 0.7095. The RSD for four samples is 133 ppm. We do not observe any systematic variations among the grape varieties (Table 1).

Strontium isotopes for bulk soil samples and their leachates, together with data for groundwater and rainwater, are presented in Table 2. The ⁸⁷Sr/⁸⁶Sr ratio of the bulk soil samples increases with depth, from 0.7138 at 30 cm to 0.7184 at 200 cm. The leached soil samples have lower isotopic ratios. However, they mirror the bulk trend of ⁸⁷Sr/⁸⁶Sr in the soil profile: the isotope ratio of the samples increases with the depth, from 0.7084 to 0.7095.

The sample of the groundwater from the well shows ⁸⁷Sr/⁸⁶Sr ~ 0.7092, almost identical to the average of the ⁸⁷Sr/⁸⁶Sr of the bio-available fraction and the average of rainwater in Central Europe (Erban Kochergina et al., 2021; Novák et al., 2023; Novak et al., 2020a; Novak et al., 2020b).

Discussion

3. Sr isotopic composition of bedrock in Southern Moravia: marine sediments, flysch sediments, granites

The geology of the studied Moravian region is very heterogeneous. In addition to loess sediments, other rocks with contrasting ⁸⁷Sr/⁸⁶Sr composition also occur in South Moravia, such as marine sediments, granitic rocks, and flysch. The whole rock Sr ratios of the marine sediments can thus be estimated from the published global marine Sr evolution curve for the Phanerozoic (McArthur et al., 2012). In general, seawater ⁸⁷Sr/⁸⁶Sr was < 0.709 during the Mesozoic and Cenozoic (McArthur et al., 2012), similar to or lower than local rainwater ⁸⁷Sr/⁸⁶Sr = 0.7097 (Novak et al., 2020a; Novak et al., 2020b).

Cretaceous to Neogene flysch sediments, characterised by a heterogeneous ⁸⁷Sr/⁸⁶Sr composition (Mason et al., 1996), range from 0.7078 to 0.7170 for the Eastern Carpathian samples. Modern ⁸⁷Sr/⁸⁶Sr of granitic rocks of the Bohemian Massif are more radiogenic than seawater carbonates, flysch, and loess sediment, with ⁸⁷Sr/⁸⁶Sr compositions of up to 0.730 (Janoušek et al., 2022; Janoušek et al., 2020). In some cases, values are extremely high (over 0.8; (Erban Kochergina et al., 2021; Hasalová et al., 2008). The heterogeneous ⁸⁷Sr/⁸⁶Sr composition of Southern Moravia bedrock allows us to test the hypothesis that such heterogeneity will be reflected in the ⁸⁷Sr/⁸⁶Sr of wines.

A recent study (Erban Kochergina et al., 2021) focused on the relationships between the Sr isotopic composition of the bedrock, water input, and plant tissue. The authors examined different tree species (oak, pine, spruce) from the contrasting geological sites without evolved soil horizons, for example, quarries and rock outcrops. The authors offered a simplified model for determining plant origin based on Sr isotopic ratios. We used this model to describe the relationship between the isotopic composition of wine, rainwater and bedrock (Figure 4). The determinative factor for this model is the relative Sr isotopic compositions of rock and rainwater. While the isotopic composition of the local rainwater varies between 0.709 and 0.710, similar to sea spray ⁸⁷Sr/⁸⁶Sr, we can divide all rocks into two groups: (i) rocks with ⁸⁷Sr/⁸⁶Sr lower than the local rainwater composition and (ii) rocks with more radiogenic ⁸⁷Sr/⁸⁶Sr than the local rainwater. The authors suggest that the plant's final ⁸⁷Sr/⁸⁶Sr composition is located on the mixing line between ⁸⁷Sr/⁸⁶Sr of rainwater and ⁸⁷Sr/⁸⁶Sr of bedrock.

The tendency of plant material to have isotopic compositions, which are a mixture of bedrock and atmospheric deposition (rain) compositions, may be used to find the origin of a wine sample from among a set of putative localities. The isotopic composition of rainfall is uniform relative to the highly variable contribution derived from the local geology. Because of this, the relative positions along the ⁸⁷Sr/⁸⁶Sr axis where the wine sample, precipitation, and bedrock of the various candidate localities plot determine whether the origin of the wine sample can be identified. This is shown in the following three cases, each involving two candidate sites: (i) The bedrock of the candidate sites falls on opposite sides of the rainfall value. Since the wine sample is a mixture of precipitation and bedrock isotopic values, the locality plotting on the opposite side of the rainfall value from the wine sample can be excluded (Figure 4A). (ii) If there are two sites with bedrock values on the same side of the rainfall value and similar to the wine sample, the wine sample could have originated from either of the two sites (Figure 4B). (iii) If there are no putative localities with bedrock on the same side of the rainfall value, there is high confidence that none of the sites is the actual origin (Figure 4B).

Figure 4. Wine origin derived from the 87Sr/86Sr systematics in bedrock and rainwater, a schematic illustration.

Save View full size Expand inline Collapse inline

The model considers simple scenarios, whereby different Sr isotope signatures of the bedrock and overlapping ranges of wine ⁸⁷Sr/⁸⁶Sr ratios due to partial uptake of atmospheric Sr (adopted from Erban Kochergina et al., 2021). This model can also be used to determine the origin of the wine. Knowing the composition of the wine and the bedrock candidates, we can determine the putative origin of the wine: (i) if the Sr isotopic signatures of bedrock/bulk soil at two candidate sites plot in opposite directions relative to the ⁸⁷Sr/⁸⁶Sr ratio of rainfall (~ 0.710), the Sr isotope analysis of the plant (wine) sample shows the probable place of origin (Figure 4A); (ii) if the ⁸⁷Sr/⁸⁶Sr ratio of the plant (wine) sample is more/less radiogenic than local rainfall, compared to the two candidate sites, it is likely that neither of the candidate sites was the place of the plant (wine) origin (Figure 4B); (iii) if the ⁸⁷Sr/⁸⁶Sr ratios of wines and two candidate sites are similar, there is no way to indicate the likely place of origin (Figure 4B).

This simplified model is confirmed by previously published results: Isotopic composition of wine from the Vulture volcanic rocks from Italy has a radiogenic Sr isotopic composition of ~ 0.708 (Marchionni et al., 2013), for example, on the mixing line between ⁸⁷Sr/⁸⁶Sr of the rock (~ 0.707), rainwater with ⁸⁷Sr/⁸⁶Sr ~ 0.710 (Rao et al., 2015), and/or seawater aerosol ~ 0.7092 (Kuznetsov et al., 2012). More radiogenic Sr in wine samples from the Portuguese wine region Douro with ⁸⁷Sr/⁸⁶Sr ~ 0.729 (Almeida and Vasconcelos, 2004) falls along the mixing line between bulk soil ⁸⁷Sr/⁸⁶Sr ~ 0.732 and atmospheric input (0.709 − 0.710).

However, numerous existing rock types have Sr isotopic composition similar to atmospheric input ~ 0.709−0.710 (Braschi et al., 2018; Epova et al., 2019; Marchionni et al., 2016). In this case, determining the wine's regional identity becomes more complicated.

3.1. Loess bedrock Sr isotopic composition

Loess is formed by clay, sand, and silt cemented by carbonates. The amount of well-soluble carbonate and its ratio to clay minerals dominate the wine's final ⁸⁷Sr/⁸⁶Sr composition. Waroszewski et al. (2021) found that samples with a higher amount of CaO and Sr content had less radiogenic ⁸⁷Sr/⁸⁶Sr; the ratio of each loess component could affect the ⁸⁷Sr/⁸⁶Sr composition of the bioavailable end member. Šimíček et al. (2021) examined loess–palaeosol sequences of Moravia and found that in loess samples, quartz minerals predominate over feldspars, phyllosilicates (mica, chlorite, clay minerals), carbonates, amphiboles, and garnets, with Ca and Sr being associated with carbonates in the medium to coarse silt and clay fractions. Due to the presence of phyllosilicates (commonly with high Rb/Sr), the bulk rock Sr isotopic ratios are likely to be high, resulting in radiogenic soil compositions, i.e., ⁸⁷Sr/⁸⁶Sr = 0.7184 in the deepest C-horizon. However, the carbonate matrix is the most enriched Sr end member in loess, and its Sr isotopic compositions are 0.7084 – 0.7095 (leached samples, Table 2). Such significant differences between bulk rock silicate and the available leached fraction are not uncommon. Similar results were obtained with carbonate (Nádaskay et al., 2019), carbonatite (Rapprich et al., 2017), and soil samples (Braschi et al., 2018; Marchionni et al., 2016; Petrini et al., 2015).

5. 87Sr/86Sr isotopic composition of Moravian wine

Almeida and Vasconcelos (2004), Tescione et al. (2020), and Marchionni et al. (2016) conclude that the winemaking process does not affect the ⁸⁷Sr/⁸⁶Sr composition of the wine. These studies also confirm that the differences in ⁸⁷Sr/⁸⁶Sr isotope ratios are the result of heterogeneity of the soil/bedrock composition and, therefore, Sr isotope composition could be a good tool to trace the origin of the wine. Taking these results into account, we were able to rule out the influence of winemaking on the ⁸⁷Sr/⁸⁶Sr isotope heterogeneity of the Moravian wines. It also allows us to compare samples from different wineries.

The ⁸⁷Sr/⁸⁶Sr of the wine samples (Figure 3) from Groups A and B are almost identical and comparable to groundwater and bio-available ⁸⁷Sr/⁸⁶Sr. The Group A and B wines belong to different sub-regions of Mikulov and Velké Pavlovice, respectively. This indicates that the geographical/historical divisions into wine (sub-) regions do not correspond to the chemical composition of wines, resulting from the locality's geological characteristics. The low heterogeneity of the wine samples is comparable to the changes in the ⁸⁷Sr/⁸⁶Sr with depth.

Group C and D samples have lower ⁸⁷Sr/⁸⁶Sr ratios than Groups A and B. However, these differences are insufficient to distinguish production from these localities.

Figure 5. 87Sr/86Sr of wines from different vintages. There are no correlations between the isotopic composition of the wine, year or grape variety. The abbreviations for varieties are listed in Table 1.

Save View full size Expand inline Collapse inline

We cannot distinguish any variations among the grape varieties (Table 1) and vintages (Figure 5). In previous studies, Durante et al. (2015) and Petrini et al. (2015) observed that the ⁸⁷Sr/⁸⁶Sr of wines of different vintages are homogeneous. Epova et al. (2019) described time-dependent changes in the Sr concentration and the ⁸⁷Sr/⁸⁶Sr ratio in wines from a winery of Pomerol AOC winery produced in 1965–2015. The authors hypothesise that this could result from geochemical processes in soils associated with Sr mobility. The slight deviation in the composition of the South Moravian wine ⁸⁷Sr/⁸⁶Sr from similar rock layers can be explained by (i) past and present soil management, (ii) the age of the vineyards, and (iii) the role of water. Water stress could play a key role in the ⁸⁷Sr/⁸⁶Sr composition of wines. We hypothesise that if plants use water from the lower horizons in dry years, the Sr radiogenic ratio could be shifted to more radiogenic (higher Sr ratios) results. This assumption can only be confirmed with a long-term dataset (e.g., Epova et al., 2019) from a winery with a long history and/or a combination of ⁸⁷Sr/⁸⁶Sr and light isotopes (O, C, H).

Based on the above, we assume that the ⁸⁷Sr/⁸⁶Sr for wines from these sub-regions of the Moravian region vary from 0.708 to 0.710, comparable to the soil leachate and groundwater composition. However, without a systematic study of the isotopic composition of wines, soils, and water from each vineyard, we cannot identify the exact origin of Southern Moravian wine. A combination with other radiogenic (e.g., Nd, Pb) or stable (O, C, H, N, S) isotopes would possibly make this possible.

6. Are Sr isotopes a good tracer for wine origin?

The indistinguishable isotopic compositions of grape juices, must, and wine enables us to see connections between them (Tescione et al., 2020). Therefore, two wine samples from the same winery, vintage, and locality should have an identical Sr isotopic composition. We can also recognise two samples from geologically different localities (e.g., basalts and granites, Figure 4). In addition, we describe the geological limitation of the connection between bedrock, soil, and wine Sr isotopic composition.

6.1. Rainwater–bedrock 87Sr/86Sr mixing line

Figure 6. Possible model of bulk rock–rainwater mixing.

Save View full size Expand inline Collapse inline

A. The diagonal line is a mixing line, for the case of wine composition = bulk rock composition and atmospheric input is negligible. The almost vertical line is another possible mixing line, where the atmospheric input is more influential. Data are from Petrini et al., 2015, Almeida and Vasconcelos, 2004; Boari et al., 2008; Braschi et al., 2018; Marchionni et al., 2013; Petrini et al., 2015; Tescione et al., 2020, Tescione et al., 2018. Horizontal arrays are the result of several wine samples being paired with the same bulk rock analysis, in many cases these are imprecise and adopted from geological maps. B. The model, linked bulk rock composition (first column) with leachate (second column) and wine (third column) composition. The last column shows that in the case of mixing samples from different geological localities, the final wine composition could vary and would not correspond to any geological locality. In this case, soils and rocks behave similarly, in the sense that the shift in isotopic ratio during the leaching process is determinative of wine isotopic ratio.

Using the published isotopic compositions of wines and bedrock, we have simulated two cases for the relationship between bedrock and wine (Figure 6A). The first situation is if the isotopic composition is similar to the bedrock. It can occur when the bedrock contains high amounts of soluble Sr, such as carbonates. The Sr input from the rainwater is insignificant due to the very low Sr content in the rainwater (Novak et al., 2020b). The opposite situation, in which the Sr isotopic composition of wine is identical to atmospheric deposition, is unrealistic since bedrock always contains a higher Sr content than rainwater. The final wine composition depends on the ratio of rainwater and soluble Sr from the rock and their isotopic compositions.

Graustein and Armstrong (1983) suggest that less than 25 % of the strontium in vegetation comes from the weathering of underlying rock and more than 75 % from atmospheric transport. Therefore, the amount of soluble Sr of the bedrock plays a key role in ⁸⁷Sr/⁸⁶Sr of wine, for example, if the bio-available Sr in the soil decreases, the role of rainwater (or irrigation) will increase.

6.2. The relationship between rainwater and the wine product

Raiber et al. 2009 compared the Sr isotopic composition of rainfall and basalt groundwater in south-eastern Australia. The freshest basalt groundwater has an ⁸⁷Sr/⁸⁶Sr signature close to the whole-rock strontium isotope ratios of the basalts (0.7045). However, down-gradient groundwater becomes more radiogenic and approaches those of local rainfall. The authors concluded that the influence of rainwater on the ⁸⁷Sr/⁸⁶Sr of the basalt groundwater is much greater than that of basaltic weathering.

6.3. Is it possible to predict 87Sr/86Sr of wine knowing its geological basement?

The model in Figure 6B describes the possible interactions between bedrock leachate and wine. It must be emphasised that no natural material on the Earth can have ⁸⁷Sr/⁸⁶Sr lower than 0.69899 (basaltic achondrite best initial BABI (Papanastassiou and Wasserburg, 1969)). Rosner (2010) pointed out that an ⁸⁷Sr/⁸⁶Sr below the modern mantle (0.702) is not possible for natural samples. The upper ⁸⁷Sr/⁸⁶Sr limit is not defined; Erban Kochergina et al. (2021) analysed granitic rock with a modern ⁸⁷Sr/⁸⁶Sr ~ 1.7.

Based on the ⁸⁷Sr/⁸⁶Sr composition, we could categorise all possible bedrocks (or bulk soils) into three groups: rocks with an isotopic composition below the rainwater composition (0.702–0.709, process I in Figure 6B), rainwater-like composition (0.709–0.710, process II in Figure 6B), and rocks with a higher isotope composition than the rainwater composition (in our model > 0.710, process III, IV, V in Figure 6B).

Rocks with an ⁸⁷Sr/⁸⁶Sr lower than 0.709 are common and are often used for cultivated wine, such as sea carbonates and basalts (Figure 6B, I trend). Such rocks have a high amount of Sr. We presume that the leachate soil fraction Sr would have isotopic ratios similar to or higher than bulk rock/soil but never higher than the local rainwater. The rocks with the lowest ⁸⁷Sr/⁸⁶Sr are ocean island basalts, such as Hawaiian basalts. These rocks have ⁸⁷Sr/⁸⁶Sr vary from 0.7033 to 0.7044 (Bryce et al., 2005), with extractable soil phases having ⁸⁷Sr/⁸⁶Sr varying from 0.7036 to ~ 0.710 (Chadwick et al., 2009). Therefore, we conclude that worldwide wine samples would have ⁸⁷Sr/⁸⁶Sr isotopic ratios higher than 0.703 and wine with such a low isotopic composition (0.703 – 0.705) is rare in continental Europe.

If the bedrock of the studied locality has ⁸⁷Sr/⁸⁶Sr close to that of the rainwater (Figure 6B, II-Trend), the bedrock and wine sample's isotopic composition is not sufficient to model the geographical origin or distinguish two samples of wine with similar composition. In this case, it is necessary to know the exact composition of rainwater and the bio-available soil fraction. However, a statistically significant number of samples and/or other analyses are still required.

Rocks with more radiogenic Sr compositions (e.g., granites) contain numerous minerals with different Sr/Rb ratios and Sr concentrations. Their differing reactivity during the weathering is determinative of the composition of the leachate and the end product, wine. There are three possible situations. Process III is widespread: the leachate has an isotopic composition between the bulk rock and rainwater (0.720 and 0.709). Another very common situation where the mineral with higher ⁸⁷Sr/⁸⁶Sr (e.g., mica with high Rb/Sr) than the bulk rock dissolves faster than other minerals (IV-process). In this case, the process described above would influence the composition of leachate and wine. Therefore, comparing wines with the bedrock composition or literature and geological map data is incorrect and may lead to wrong conclusions. The V-process was described by Blum et al., (2002), who showed that plants' root mycorrhizal systems can directly affect apatite weathering. Apatite is a common Rb-free mineral, so its isotopic composition is constant through geologic time compared to other minerals (feldspars, mica) and can be below 0.709. During the V-process, the plant could have a lower ⁸⁷Sr/⁸⁶Sr than the rainwater, whereas the bulk rock has a more radiogenic ⁸⁷Sr/⁸⁶Sr composition, which is higher than the rainwater. In the example described by Blum et al. (2002), the apatite has a ⁸⁷Sr/⁸⁶Sr composition between that of silicate rock and atmospheric deposition, and, therefore, it is C-process in our model. We suggest that the V-process is very unlikely for wine production. However, we cannot rule out this unrealistic situation.

If fruit from different sites with different bedrocks were used during the winemaking process, the blended product could have a wide range in its ⁸⁷Sr/⁸⁶Sr isotopic composition. Hence, all models and conclusions would be incorrect. In this case, the grape and grape juice analyses could be used as end members in the winemaking process modelling.

Conclusion

This paper presents the first and original ⁸⁷Sr/⁸⁶Sr data on wine from the Moravian region in the Czech Republic. We compare its isotopic composition with that of bulk soil and its bio-available soil fraction from 30, 60, 90, and 200 cm depths.

The studied soils are heterogeneous in their ⁸⁷Sr/⁸⁶Sr composition; bulk soil composition varies from 0.7138 to 0.7222. The bio-available soil fraction varies from 0.7082 to 0.7105, with the average composition similar to the groundwater from the local well. We conclude that the groundwater sample is a good tracer for bio-available Sr.

We offer a simple model that connects the wine-leachate-bulk rock/soil Sr isotopic composition. We recommend analysing the rainwater and groundwater and leachate from soil together with wine samples. We show that there are situations in which the composition of the bulk rock can differ significantly from that in the wine. In the case of bedrocks with ⁸⁷Sr/⁸⁶Sr below the rainwater composition, we do not need to provide the soil leaching tests. In this case, the isotopic composition of wine would vary from 0.703 to rainwater ⁸⁷Sr/⁸⁶Sr.

The ⁸⁷Sr/⁸⁶Sr for wines from South Moravia with sedimentary bedrocks vary from 0.708 to 0.710, which correlates to the ⁸⁷Sr/⁸⁶Sr in soil leachate. We do not observe substantial differences within and between the cultivars and vintages. Due to the similarity in the isotopic composition of the bedrocks of studied vineyards, we cannot distinguish between different wine sub-regions of Southern Moravia.

Acknowledgements

We thank L. Šlancar, wine producer from Ludwig, s. r. o. Czech Republic for logistical and financial support and H. Maršíčková (CGS Prague) for technical assistance in the isotopic laboratory, as well as an anonymous reviewer and the editors for their insightful comments. This work was supported by the Center for Landscape and Biodiversity (DivLand) (Technology Agency of the Czech Republic (TAČR) SS02030018).

References

• Albaréde, F. (1995). Introduction to geochemical modeling. Cambridge, New York, Port Chester, Melbourne, Sydney: Cambridge University Press. 543 pp.
• Almeida, C. M. R., & Vasconcelos, M. T. S. (2004). Does the winemaking process influence the wine 87Sr/86Sr? A case study. Food Chemistry, 85(1), 7–12. https://doi.org/10.1016/j.foodchem.2003.05.003
• Blum, J. D., Klaue, A., Nezat, C. A., Driscoll, C. T., Johnson, C. E., Siccama, T. G., Eagar, C., Fahey, T. J., & Likens, G. E. (2002). Mycorrhizal weathering of apatite as an important calcium source in base-poor forest ecosystems. Nature, 417(6890), 729–731. https://doi.org/10.1038/nature00793
• Boari, E., Tommasini, S., Mercurio, M., Morra, V., Mattei, M., Mulinacci, N., & Conticelli, S. (2008). 87Sr/86Sr of some Central and Southern Italian wines and its use as fingerprints for Geographic Provenance. Proceedings of 31st World Congress of Vine and Wine, June, 1–6.
• Braschi, E., Marchionni, S., Priori, S., Casalini, M., Tommasini, S., Natarelli, L., Buccianti, A., Bucelli, P., Costantini, E. A. C., & Conticelli, S. (2018). Tracing the 87Sr/86Sr from rocks and soils to vine and wine: An experimental study on geologic and pedologic characterisation of vineyards using radiogenic isotope of heavy elements. Science of The Total Environment, 628–629, 1317–1327. https://doi.org/10.1016/j.scitotenv.2018.02.069
• Bryce, J. G., DePaolo, D. J., & Lassiter, J. C. (2005). Geochemical structure of the Hawaiian plume: Sr, Nd, and Os isotopes in the 2.8 km HSDP-2 section of Mauna Kea volcano. Geochemistry, Geophysics, Geosystems, 6(9), 1–36. https://doi.org/10.1029/2004GC000809
• Chadwick, O. A., Derry, L. A., Bern, C. R., & Vitousek, P. M. (2009). Changing sources of strontium to soils and ecosystems across the Hawaiian Islands. Chemical Geology, 267(1–2), 64–76. https://doi.org/10.1016/j.chemgeo.2009.01.009
• Crittenden, R. G., Andrew, A. S., LeFournour, M., Young, M. D., Middleton, H., & Stockmann, R. (2007). Determining the geographic origin of milk in Australasia using multi-element stable isotope ratio analysis. International Dairy Journal, 17(5), 421–428. https://doi.org/10.1016/j.idairyj.2006.05.012
• Durante, C., Baschieri, C., Bertacchini, L., Bertelli, D., Cocchi, M., Marchetti, A., Manzini, D., Papotti, G., & Sighinolfi, S. (2015). An analytical approach to Sr isotope ratio determination in Lambrusco wines for geographical traceability purposes. Food Chemistry, 173, 557–563. https://doi.org/10.1016/j.foodchem.2014.10.086
• Epova, E. N., Bérail, S., Séby, F., Vacchina, V., Bareille, G., Médina, B., Sarthou, L., & Donard, O. F. X. (2019). Strontium elemental and isotopic signatures of Bordeaux wines for authenticity and geographical origin assessment. Food Chemistry, 294(April), 35–45. https://doi.org/10.1016/j.foodchem.2019.04.068
• Erban Kochergina, Y. V., Erban, V., & Hora, J. M. (2022). Sample preparation and chromatographic separation for Sr, Nd, and Pb isotope analysis in geological , environmental , and archaeological samples. Journal of Geosciences, 67, 259–271. https://doi.org/10.3190/jgeosci.357
• Erban Kochergina, Y. V., Novak, M., Erban, V., & Stepanova, M. (2021). 87Sr/86Sr isotope ratios in trees as an archaeological tracer: Limitations of linking plant-biomass and bedrock Sr isotope signatures. Journal of Archaeological Science, 133(June), 105438. https://doi.org/10.1016/j.jas.2021.105438
• Graustein, W. C., & Armstrong, R. L. (1983). The Use of Strontium-87/Strontium-86 Ratios to Measure Atmospheric Transport into Forested Watersheds. Science, 219(4582), 289–292. https://doi.org/10.1126/science.219.4582.289
• Hasalová, P., Janoušek, V., Schulmann, K., Štípská, P., & Erban, V. (2008). From orthogneiss to migmatite: Geochemical assessment of the melt infiltration model in the Gföhl Unit (Moldanubian Zone, Bohemian Massif). Lithos, 102(3–4), 508–537. https://doi.org/10.1016/j.lithos.2007.07.021
• Horn, P., Schaaf, P., Holbach, B., Hölzl, S., & Eschnauer, H. (1993). 87Sr/86Sr from rock and soil into vine and wine. Zeitschrift Für Lebensmittel-Untersuchung Und -Forschung, 196(5), 407–409. https://doi.org/10.1007/BF01190802
• Janoušek, V., Erban Kochergina, Y. V., Andronikov, A.V., & Kusbach, V. (2022). Decoupling of Mg from Sr–Nd isotopic compositions in Variscan subduction-related plutonic rocks from the Bohemian Massif: implications for mantle enrichment processes and genesis of orogenic ultrapotassic magmatic rocks. International Journal of Earth Sciences, 111, 1491–1518. https://doi.org/10.1007/s00531-022-02199-w
• Janoušek, V., Hanžl, P., Svojtka, M., Hora, J. M., Erban Kochergina, Y. V., Gadas, P., Holub, F. V., Gerdes, A., Verner, K., Hrdličková, K., Daly, J. S., Buriánek, D., Kochergina, Y. V. E., Gadas, P., Holub, F. V., Gerdes, A., Verner, K., Hrdličková, K., Daly, J. S., & Buriánek, D. (2020). Ultrapotassic magmatism in the heyday of the Variscan Orogeny: the story of the Třebíč Pluton, the largest durbachitic body in the Bohemian Massif. International Journal of Earth Sciences, 110(5), 1767–1810. https://doi.org/10.1007/s00531-021-01993-2
• Janoušek, V., Moyen, J., Martin, H., Erban, V., & Farrow, C. (2016). Geochemical Modelling of Igneous Processes – Principles And Recipes in R Language. Dans Springer eBooks. https://doi.org/10.1007/978-3-662-46792-3
• Jochum, K. P., & Nohl, U. (2008). Reference materials in geochemistry and environmental research and the GeoReM database. Chemical Geology, 253(1–2), 50–53. https://doi.org/10.1016/j.chemgeo.2008.04.002
• Kment, P., Mihaljevič, M., Ettler, V., Šebek, O., Strnad, L., & Rohlová, L. (2005). Differentiation of Czech wines using multielement composition – A comparison with vineyard soil. Food Chemistry, 91(1), 157–165. https://doi.org/10.1016/j.foodchem.2004.06.010
• Kochergina, Y. V, Udatný, M., Penížek, V., & Mihaljevič, M. (2017). Mobility of Pb, Zn, Cu and As in disturbed forest soils affected by acid rain. Environmental Monitoring and Assessment, 189(11), 570. https://doi.org/10.1007/s10661-017-6306-7
• Kuznetsov, A. B., Semikhatov, M. A., & Gorokhov, I. M. (2012). The Sr isotope composition of the world ocean, marginal and inland seas: Implications for the Sr isotope stratigraphy. Stratigraphy and Geological Correlation, 20(6), 501–515. https://doi.org/10.1134/S0869593812060044
• Lisá, L., & Uher, P. (2006). Provenance of Würmian loess and loess-like sediments of Moravia and Silesia (Czech Republic): A study of zircon typology and cathodoluminiscence. Geologica Carpathica, 57(5), 397–403.
• Marchionni, S., Braschi, E., Tommasini, S., Bollati, A., Cifelli, F., Mulinacci, N., Mattei, M., & Conticelli, S. (2013). High-precision 87Sr/86Sr analyses in wines and their use as a geological fingerprint for tracing geographic provenance. Journal of Agricultural and Food Chemistry, 61(28), 6822–6831. https://doi.org/10.1021/jf4012592
• Marchionni, S., Buccianti, A., Bollati, A., Braschi, E., Cifelli, F., Molin, P., Parotto, M., Mattei, M., Tommasini, S., & Conticelli, S. (2016). Conservation of 87Sr/86Sr isotopic ratios during the winemaking processes of ‘Red’ wines to validate their use as geographic tracer. Food Chemistry, 190, 777–785. https://doi.org/10.1016/j.foodchem.2015.06.026
• Mason, P. R. D., Downes, H., Thirlwall, M. F., Seghedi, I., Szakács, A., Lowry, D., & Mattey, D. (1996). Crustal Assimilation as a Major Petrogenetic Process in the East Carpathian Neogene and Quaternary Continental Margin Arc, Romania. Journal of Petrology, 37(4), 927–959. https://doi.org/10.1093/petrology/37.4.927
• McArthur, J. M., Howarth, R. J., & Shields, G. A. (2012). Strontium Isotope Stratigraphy. In The Geologic Time Scale (pp. 127–144). Elsevier. https://doi.org/10.1016/B978-0-444-59425-9.00007-X
• Moreira, C., de Pinho, M., Curvelo-Garcia, A., de Sousa, B., Ricardo-da-Silva, J., & Catarino, S. (2017). Evaluating Nanofiltration Effect on Wine 87Sr/86Sr Isotopic Ratio and the Robustness of this Geographical Fingerprint. South African Journal of Enology & Viticulture, 38(2), 82–93. https://doi.org/10.21548/38-1-942
• Nádaskay, R., Kochergina, Y. V. E., Čech, S., Švábenická, L., Valečka, J., Erban, V. V., Halodová, P., & Čejková, B. (2019). Integrated stratigraphy of an offshore succession influenced by intense siliciclastic supply: Implications for Coniacian tectono-sedimentary evolution of the West Sudetic area (NW Bohemian Cretaceous Basin, Czech Republic). Cretaceous Research, 102, 127–159. https://doi.org/10.1016/j.cretres.2019.06.005
• Novák, M., Andronikov, A. V., Holmden, C., Erban Kochergina, Y. V., Veselovský, F., Pačes, T., Vítková, M., Kachlík, V., Šebek, O., Hruška, J., Štěpánová, M., Čuřík, J., Přechová, E., Fottová, D., Andronikova, I. E., Erban, V., Koubová, M., Vostrá, I., Housková, M., & Komárek, A. (2023). δ26Mg, δ44Ca and 87Sr/86Sr isotope differences among bedrock minerals constrain runoff generation in headwater catchments: An acidified granitic site in Central Europe as an example. CATENA, 221, 106780. https://doi.org/10.1016/j.catena.2022.106780
• Novak, M., Holmden, C., Farkaš, J., Kram, P., Hruska, J., Curik, J., Veselovsky, F., Stepanova, M., Kochergina, Y. V., Erban, V., Fottova, D., Simecek, M., Bohdalkova, L., Prechova, E., Voldrichova, P., & Cernohous, V. (2020a). Calcium and strontium isotope dynamics in three polluted forest ecosystems of the Czech Republic, Central Europe. Chemical Geology, 536(June 2019), 119472. https://doi.org/10.1016/j.chemgeo.2020.119472
• Novak, M., Holmden, C., Farkas, J., Kram, P., Hruska, J., Curik, J., Veselovsky, F., Stepanova, M., Kochergina, Y. V. E., Erban, V., Andronikov, A., Sebek, O., Koubova, M., Bohdalkova, L., & Vitkova, H. (2020b). Magnesium and calcium isotope systematics in a headwater catchment underlain by amphibolite: Constraints on Mg–Ca biogeochemistry in an atmospherically polluted but well-buffered spruce ecosystem (Czech Republic, Central Europe). CATENA, 193, 104637. https://doi.org/10.1016/j.catena.2020.104637
• Papanastassiou, D. A., & Wasserburg, G. J. (1969). Initial strontium isotopic abundances and the resolution of small time differences in the formation of planetary objects. Earth and Planetary Science Letters, 5, 361–376. https://doi.org/10.1016/S0012-821X(68)80066-4
• Petrini, R., Sansone, L., Slejko, F. F., Buccianti, A., Marcuzzo, P., & Tomasi, D. (2015). The 87Sr/86Sr strontium isotopic systematics applied to Glera vineyards: A tracer for the geographical origin of the Prosecco. Food Chemistry, 170, 138–144. https://doi.org/10.1016/j.foodchem.2014.08.051
• Rao, W., Han, G., Tan, H., & Jiang, S. (2015). Chemical and Sr isotopic compositions of rainwater on the Ordos Desert Plateau, Northwest China. Environmental Earth Sciences, 74(7), 5759–5771. https://doi.org/10.1007/s12665-015-4594-1
• Rapprich, V., Kochergina, Y. V. E., Magna, T., Laufek, F., Halodová, P., & Bůzek, F. (2017). Carbonate-rich dyke in Roztoky Intrusive Complex - an evidence for carbonatite magmatism in the Eger Rift? Journal of Geosciences, 62(2), 121–136. https://doi.org/10.3190/jgeosci.238
• Rosner, M. (2010). Geochemical and instrumental fundamentals for accurate and precise strontium isotope data of food samples: Comment on “Determination of the strontium isotope ratio by ICP-MS ginseng as a tracer of regional origin” (Choi et al., 2008). Food Chemistry, 121(3), 918–921. https://doi.org/10.1016/j.foodchem.2010.01.019
• Rummel, S., Hoelzl, S., Horn, P., Rossmann, A., & Schlicht, C. (2010). The combination of stable isotope abundance ratios of H, C, N and S with 87Sr/86Sr for geographical origin assignment of orange juices. Food Chemistry, 118(4), 890–900. https://doi.org/10.1016/j.foodchem.2008.05.115
• Santos, J. A., Fraga, H., Malheiro, A. C., Moutinho-Pereira, J., Dinis, L.-T., Correia, C., Moriondo, M., Leolini, L., Dibari, C., Costafreda-Aumedes, S., Kartschall, T., Menz, C., Molitor, D., Junk, J., Beyer, M., & Schultz, H. R. (2020). A Review of the Potential Climate Change Impacts and Adaptation Options for European Viticulture. Applied Sciences, 10(9), 3092. https://doi.org/10.3390/app10093092
• Schaller, K. (2017). Terroir - Myth and/or Reality - Outstanding Marketing Idea? A Review. Notulae Botanicae Horti Agrobotanici Cluj-Napoca, 45(2), 332–342. https://doi.org/10.15835/nbha45210898
• Šimíček, D., Bábek, O., Hron, K., Pavlů, I., & Kapusta, J. (2021). Separating provenance and palaeoclimatic signals from particle size and geochemistry of loess-palaeosol sequences using log-ratio transformation: Central European loess belt, Czech Republic. Sedimentary Geology, 419, 105907. https://doi.org/10.1016/j.sedgeo.2021.105907
• Tescione, I., Casalini, M., Marchionni, S., Braschi, E., Mattei, M., & Conticelli, S. (2020). Conservation of 87Sr/86Sr During Wine-Making of White Wines: A Geochemical Fingerprint of Geographical Provenance and Quality Production. Frontiers in Environmental Science, 8 (September), 1–13. https://doi.org/10.3389/fenvs.2020.00153
• Tescione, I., Marchionni, S., Casalini, M., Vignozzi, N., Mattei, M., & Conticelli, S. (2018). 87Sr/86Sr isotopes in grapes of different cultivars: A geochemical tool for geographic traceability of agriculture products. Food Chemistry, 258(March), 374–380. https://doi.org/10.1016/j.foodchem.2018.03.083
• Waroszewski, J., Pietranik, A., Sprafke, T., Kabała, C., Frechen, M., Jary, Z., Kot, A., Tsukamoto, S., Meyer-Heintze, S., Krawczyk, M., Łabaz, B., Schultz, B., & Erban Kochergina, Y.V. (2021). Provenance and paleoenvironmental context of the Late Pleistocene thin aeolian silt mantles in southwestern Poland – A widespread parent material for soils. CATENA, 204(October 2020), 105377. https://doi.org/10.1016/j.catena.2021.105377