Svaneti and Racha-Lechkumi are historical provinces of Georgia, located on the south-facing macro-slope of the western part of the Greater Caucasus. Svanetis Kavkasioni, a part of the Main Watershed Range of the Greater Caucasus, creates the natural northern border of Svaneti between Gvandra and Mt. Pasismta. From the west the region is bordered by the Kodori and Akibo ranges, from the south-west by the rivers Bordjali and Gandishis-Ghele, both tributaries of the river Enguri, as well as by the Bokunstha range. From the east Svaneti is bordered by the northern part of Lechkhumi range, and from the south by the southern part of the same Lechkhumi range, and the eastern part of Egrisi range. The Svaneti range divides the region into two depressions: Zemo (Upper) Svaneti and Kvemo (Lower) Svaneti, creating a watershed between the Enguri Tskhenistskali basins. The region has an altitudinal gradient from 800-4,500 masl [1,2]. Administratively, Zemo Svaneti belongs to Mest’ia municipality (with an area of 3,045 km2), which is part of the Samegrelo-Zemo Svaneti Mkhare (region) and includes Mest’ia, as its administrative center, and 134 villages of 16 communes: Becho, Ch’uberi, Etseri, Idliani, Ipari, Kala, Khaishi, Latali, Lakhamula, Lenjeri, Mulakhi, Nakra, Pari, Tskhumari, Ts’virmi, and Ushguli. Kvemo Svaneti is part of Lentekhi municipality (with an area of 1,344 km²) and belongs to the Racha-Lechkhumi-Kvemo Svaneti Mkhare. It includes the settlement Lentekhi as administrative center, and 59 villages of seven communes: Chikhareshi, Choluri, Kheledi, Khopuri, Rtskhmeluri, Tsana, and Zhakhunderi [3].

The area, as part of the western Greater Caucasus, experiences the influence of the Black Sea and moist westerly winds. The mean temperature of the warmest months (July-August) in Svaneti decreases from +22°C at relatively low altitudes to +7°C to -1°C above 3,200 masl; and the mean temperature of the coldest month (January) from +10°C to -30°C or -35°C. The annual precipitation is 1,500-2,000 mm [4].

The vegetation of the region includes montane forest, subalpine, alpine, subnival and nival zones and corresponds to the West Caucasian, i.e., Colchic, type of the vegetation vertical zonation [1,5,6].

Most inhabitants speak both Georgian and Svanetian. The Svanetian language, like Georgian, belongs to the Kartvelian group of the Iberian-Caucasian family of languages. Svanetian is considered an old branch, and its archaisms considered very important in Proto-Kartvelian linguistics, but Svanetian still shares common basic vocabulary and grammar structures with Georgian. Svanetian does not have written language - historically the common colloquial language for all the Georgian tribes as well as formal language in the Georgian State(s) was Georgian - but rich Svanetian oral folklore provides a sizable corpus. The Svanetian language has four dialects: Balszemouri and Balskvemouri in Zemo Svaneti, and Lashkhuri and Lentekhuri in Kvemo Svaneti. In Kvemo Svaneti some researchers distinguish a third dialect, Choluruli, which is very close to Lashkhuri. Each of the listed dialects is only used in small groups of villages in Svaneti [7-15].

Svaneti has always been one of the more accessible mountain regions of Georgia, and recently winter tourism has experienced a boom. However, surprisingly few studies on the plant use of its inhabitants exist [16-18].

In this study we documented traditional plant use in Svvaneti and Racha-Lechkhumi, and hypothesized that (1) plant use knowledge in general would be higher in isolated high elevation communities, and that (2) use of home gardens would be much more restricted to lower elevation settings.

Study area

Administratively, Zemo Svaneti belongs to Mest’ia municipality (with an area of 3,045 km²), which is part of the Samegrelo-Zemo Svaneti Mkhare (region) and includes Mest’ia, as its administrative center, and 134 villages of 16 communes: Becho, Ch’uberi, Etseri, Idliani, Ipari, Kala, Khaishi, Latali, Lakhamula, Lenjeri, Mulakhi, Nakra, Pari, Tskhumari, Ts’virmi, and Ushguli. Kvemo Svaneti is part of Lentekhi municipality (with an area of 1,344 km²) and belongs to the Racha-Lechkhumi-Kvemo Svaneti Mkhare. It includes the settlement Lentekhi as administrative center, and 59 villages of seven communes: Chikhareshi, Choluri, Kheledi, Khopuri, Rtskhmeluri, Tsana, and Zhakhunderi [15].

The area, as part of the western Greater Caucasus, experiences the influence of the Black Sea and moist westerly winds. The mean temperature of the warmest months (July-August) in Svaneti decreases from +22°C at relatively low altitudes to +7°C to -1°C above 3,200 masl; and the mean temperature of the coldest month (January) from +10°C to -30°C or -35°C. The annual precipitation is 1,500-2,000 mm [16].

The vegetation of the region includes montane forest, subalpine, alpine, subnival and nival zones and corresponds to the West Caucasian, i.e., Colchic, type of the vegetation vertical zonation [1,17,18].

Ethnobotanical interviews

Fieldwork was conducted in Svaneti and Racha July-August 2014 in 17 communities. Interviews using semi-structured questionnaires were conducted with 63 participants (29 women and 34 men) with oral prior informed consent. The participants were selected by snowball sampling, trying to reach gender balance and represent members of different age (13-91 years). However, most participants were over 50 years old, because only very few younger people remain in remote Georgian villages. All interviews were carried out in the participants’ homes and gardens by native speakers of Georgian and Svan dialects, and then translated into English. Plants grown in the home-gardens were used as prompts, while wild-collected species were free listed. In contrast to 5 many other countries Georgia may refer to a complete flora [19-21], as well as a broad inventory of vernacular names in all languages [22]. Wild-collected and home-garden species were identified in the field using this literature [19-21], and vouchers collected and deposited in the National Herbarium of Georgia. The nomenclature of all species follows www.tropicos.org, under APGIII [23].

Statistical analysis

In order to characterize similarities and differences between individual knowledge sets, the distance among informants was calculated using non-metric multi-dimensional scaling on two distance matrices: one in which columns represented plant species reported, and one in which columns represented uses reported. The resulting ordinations plot individuals who report similar plants or similar uses more closely together. We then fit different environmental vectors (community elevation) and factors (gender, community) to test how whether a characteristic explains the location of informants in the ordination. We compared these fits to 999 randomized shuffles of the environmental variables to calculate significance. We calculated informant consensus (IC) for a given Use Category was calculated as the number of use reports minus the number of taxa over the number of use reports minus one. Species were ranked by three metrics: Cultural Importance Value (CIV), the sum within species across all plant-uses of the number of informants reporting a plant-use over the number of informants reporting the plant; Use Diversity (UD), the Shannon Index of uses (calculated with the R package vegan [24]; and Use Value (UV), the number of reports of a species over total number of informants asked in a region [25]. For these analyses, we considered two levels of geographic structure - the community of 6 interview, and four larger geographic areas: Racha Lechkhumi - Tsageri (16 informants in 2 communties), Svaneti-High (20 informants in 6 communities), Svaneti-Low (20 informants in 7 communities), and Svaneti-Ushguli (7 informants in 2 communities).

We encountered 203 plant species belonging to 144 genera of 65 families being used in the research region (Table 1). Of these, 99 species were exclusively wild-collected, 73 were grown in home-gardens, and 35 were both grown in home-gardens and collected in the wild. Plants and their uses mostly overlapped among the four areas within the region, with a slightly wider divergence in uses than in plants (Figures 1-3). The ordination in use-space showed greater convergence on the use categories construction, utensils and tools, and food; while some clusters of informants were differentiated by reporting relatively more medicinal, veterinary, fuel or cultural uses. The environmental fit analysis showed that a large degree of this variation was explained by differences among informant communities (Figure 4, r²=0.84, p=0.001). Community also significantly fit the ordination for both plantspace (r²=0.78, p=0.001). The elevation of the informant community significantly fit the ordination in plant-space and explained a large degree of the variation in plant species reported (Figure 5A, r²=0.65, p=0.001) but not in use-space (Figure 5B, r²=0.02, p=0.51). Gender was not significant in plant-space (p=0.81) or use-space (p=0.75).

Figure 1: Study area (based on United Nations, modified).

Figure 2: Proportional Euler diagrams of plants and usage shared among areas within Svaneti-Racha.

Figure 3: Informants from the Svaneti-Racha region ordered by their distance in uses reported, labelled by identification number (see Table 1).

Figure 4: Community explains a large degree of the variation in uses reported (r²=0.84, p=0.001).

Figure 5: Elevation of community explants a large degree of the variation in plant species reported (A, r²=0.65, p=0.001) but not in uses reported (B, r²=0.02, p=0.51).

Informant consensus

Relative number of use reports and informant consensus (IC) across communities showed some similar patterns across different use categories (Figures 5 and 6; Table 2). Number of uses reports tended 7 to be highest in the food use category, and IC increased with increasing use reports. Diverging from this pattern, Leli, with a relatively small number of use reports, showed a much higher IC, and was the only community to report cultural uses. The medicinal and utensils and tools use categories exhibited much greater IC, even at low numbers of use reports, in comparison to, e.g., the construction and food use categories.

Figure 6: Informant consensus plotted over number of use reports for each Use Category among informant communities. Micronucleus induced in mice exposed to TiO₂-NPs. (a) PCE: Polychromatic erythrocytes, NCE: Normochromatic erythrocyte (b) MNPCE: mono-micronucleated polychromatic erythrocyte, (c) Bi-MNPCE: Bi- micronucleated polychromatic erythrocyte.

Table 2: Mean informant consensus across use categories among informant communities, with total number of use reports and taxa.

Plant relative importance

The three different plant species importance ranking metrics produced quite different rankings of plant importance. Cultural Importance (CI, Table 3) prioritized species of diverse life-forms and use categories, but species with the very highest CI were not often those with high Use Diversity (UD) or Use Value (UV). Corylus avellana L./C. pontica K. Koch is an exception, with high importance in all three metrics, and Sambucus ebulus L. and Junperus sabina L. ranked highly in both CI and UD. These two species, and others with especially high UD (Table 4) tended to be be woody species. Species with the greatest UD did tend to also have high CI or UV metrics. Species with high Use Value (UV, Table 5) tended to be common managed/domesticated species from home orchards, gardens or farms.

Table 3: The 95th percentile species ranked by Cultural Importance.

Table 4: The 95th percentile species ranked by Use Diversity.

Table 5: The 95th percentile species ranked by Use Value.

For Svaneti-Racha, geographic differences were important for in which kinds of ways people used plants; and in which plants they used. Elevation also played an important role in what plants they use, but not in use categories. Medicinal and utensils and tools use categories exhibited much greater IC at low numbers of use reports Cultural Importance (CI, Table 3) prioritized species of diverse life-forms and use categories but looking at the list, is there may be a tendency for aromatic/bitter profiles? (compare with, for instance, Table 5).

We found Corylus avellana L./C. pontica K. Koch to be one of the most important useful plants in the region by any of the three metrics we use. Hazelnuts are an excellent example of Georgia’s place as a hotspot of temperate plant diversity and an ancient center of plant domestication, home to a diversity of endemic crop varieties and ecotypes. Hazelnuts’ earliest cultivation was in the Caucasus, and the few current studies on the topic suggest that Georgian hazelnuts have a remarkable degree of genetic diversity. This botanical richness is held in small farms like those in the study area: 80% of Gerogian hazelnut producers cultivate orchards of <0.5 hectare (Lane 2013). The importance of hazelnuts in the area implies a rich and body of plant genetic resources and associated knowledge.

Plant species, and uses, found in our study, showed clear relations to the wider Eurasian cultural complex, showing broad overlap with other studies, forming part of what Biscotti and Pieroni [26] described as “hidden Mediterranean diet”. The number of species used in Svaneti and Racha was however far higher than in any published study from either the region or the wider Eurasia [27-39]. The much larger incidence of plant use for food might stem from the particularly long agricultural and gardening history in the Caucasus. Interestingly, medicinal plant species tended to coincide more with other studies in the region [27,40-45].

High UV scores are found in garden rather than in wild-collected species, e.g., Malus domestica, Pyrus communis, Coriandrum sativum, Corylus avellana / C. pontica, Allium victorialis, Vitis vinifera; and garden species also differed much less across environmental gradients than wildcollected species. To some extent, this may reflect the wide geographic distribution of these cultivated species and their cultivated niche, in contrast to wilder areas with more niche divergence along geographical and topographical gradients. It also underlines the importance of Georgia as an ancient center of crop domestication and diversification and its role in the origin of many domesticated species that are globally spread today.

The wide use of mushrooms and lichens in Svaneti is of particular interest, and characterizes the region as mycophilous [46].

While the great variety of plant species used in the Georgian Caucasus might provide a reservoir for food security, similar to the Balkans [47], climate change is starting to affect both natural floristic diversity and gardens both in the Caucasus as well as continent wide [48,49].

For Svaneti-Racha, geographic differences were important for in which kinds of ways people used plants, and which plants they used. Elevation also played an important role in what plants were utilized, but not in use categories. Medicinal and utensils and tools use categories exhibited much greater IC at low numbers of use reports, while Cultural Importance prioritized species of diverse life-forms and use categories.

The shift from ancient cultivars to modern high-yielding crops such as maize and potato, which took place in the lowland areas much earlier, began in mountain villages towards the end of Soviet occupation. Similar changes have been reported from other former Soviet republics [50]. In addition, the rehabitation of high altitude villages has been only partial - while some families have returned at least for the summer, many villages remain in ruins. In occupied villages old household utensils like butter barrels are often to be found in storage, but not used anymore. Small bridges are still made from wood, but many other wooden household items like beautiful bedheadboards are simply discarded. Some implements, e.g., snowshoes or brooms are still maintained. Agricultural tools such as hay rakes are a common sight in abandoned barns, but more sought afar 10 items like ox-drawn threshing sledges could only be found in museums. While sheep were produced on a large scale during Soviet times, leading to widespread overgrazing, nowadays only a few scattered herds remain, and traditional wool items are getting more difficult to find, while tourist products abound along roadsides especially in the outskirts of Tbilisi and resort areas like Borjomi and Barisako. Sadly, we could only find some cultivation of Hordeum in Svaneti, although many participants mentioned that old landraces of wheat and barley were formerly preferred to prepare bread and beer for religious rituals. All over Georgia abandoned terraces indicate where grain was formerly grown. Many old barns still contain clay lined grain storage baskets made from Salix sp., which quite often contain old grains. One old storage chest in an abandoned barn was still half full of oats, probably harvested in the 1970s, and some wheat bran was still found in an abandoned house. Now villagers buy wheat to distil alcohol or to bake bread, or buy commercial beer making mixtures to brew their own beer [17].

The maintenance of home gardens in Georgia serves as socioecological memory, like in other regions [51,52], and as such is an irreplaceable tool to maintain Georgian culture. In contrast to other regions, this represents not just a reflectance of growing popularity of gardening and gathering [53-60], but cultural survival.

We would like to thank and all the people of Svaneti and Racha-Lechkhumi who kindly agreed to participate in the ethnobiological interviews.

Before conducting interviews, prior informed consent was obtained from all participants. No further ethics approval was required.

This manuscript does not contain any individual person’s data and further consent for publication is not required.

The raw data contain the names of all participants, and cannot be shared in this form. All other material is included in the paper.

This study was funded through William L. Brown Center endowment funds. The funding body itself has no direct role in the design of the study, collection or analysis of the data and use of results.

The authors declare that they have no competing interests.

RWB, NYPZ, SS, ZK, DK, DT and KB designed the study; RWB, NYPZ, SS, ZK, DT and KB conducted the fieldwork, RHE conducted the main statistical analysis; RBU, NYPZ and RHE analyzed the data and wrote the manuscript; all authors read, corrected and approved the manuscript.