Introduction
Tracking the material evidence for a Greek military presence between 329 and 327 BC—Alexander of Macedon’s campaign—and at the end of the third century BC—the campaign of Antiochus III—is a perennial problem in the archaeology of Central Asia. Although the written sources give accounts of battles and the advance of troops and provide partial descriptions of the locations in which these events took place, it is rarely possible to directly link battles with specific places and surviving monuments. For this reason, the ‘Darband Wall’, one of Central Asia’s most distinctive ancient linear fortifications, draws considerable scholarly interest. Discovered in 1986 and located between the villages of Darband and Shurob, in the Baysun District of the southern Uzbekistan province of Surxondaryo (Figure 1), the wall marks the presumed border between the historic regions of Bactria and Sogdiana. It was generally believed that the wall was a part of measures to protect Bactria from nomadic incursions during the war between Antiochus III of the Seleucids and the Greco-Bactrian king Euthydemus I (c. 208–206 BC) (e.g. Stančo Reference Stančo2021: 87–88).
Figure 1. Location of the Darband Wall and the fortress of Uzundara in the Baysun District, south Uzbekistan (map by L. Stančo).
Bactria and Sogdiana (Bakhdi and Suguda in Old Persian) were inhabited by eastern Iranian peoples under the rule of the Achaemenid Empire (c. 550–330 BC), but both came under Greek control during the campaign of Alexander of Macedon (in 329 and 327 BC, respectively), and, in the case of Bactria, remained that way for two centuries. Bactria covered the upper reaches of the Amu Darya (Oxus) River, while Sogdiana lay in the basin of the Kashka Darya and the Zerafshan rivers, extending as far as the Syr Darya (Jaxartes). The exact boundaries of these regions remain a subject of scholarly debate and may have historically shifted between the Baysun-Kugitang Mountains and the Amu Darya River (Lyonnet Reference Lyonnet and Mairs2020: 313); the attribution of the Surkhan Darya and Kafirnigan valleys, in particular, is uncertain. To acknowledge this uncertainty, we use the term Bactro-Sogdian borderlands, though, of course, the discussion of borders has implications for the topic addressed here.
Field excavations prompted interpretations of the function and dating of the Darband Wall, although the work conducted by two archaeological teams differed in their conclusions (Rtveladze Reference Rtveladze1986; Rakhmanov Reference Rakhmanov and Masson1994). Despite intensive research at the turn of the millennium, the findings were not published internationally until two decades later (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022) and actual evidence for the dating of this structure remains scarce. The largest and best-preserved sections of defensive architecture were dated (based on their appearance) to the Kushan period (c. first–third centuries AD) (Rtveladze Reference Rtveladze1986: 35, Reference Rtveladze1990: 9–11), with only the oldest and least significant phase of construction suggested to date to the Greco-Bactrian period (c. 250–140 BC) (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 241). The present study aims to verify this chronology, determining the most accurate dating of individual phases of the structure through a series of new radiocarbon dates, and to offer a plausible basis for further historical interpretations of this important fortification. We focus primarily on refining the dating of the Darband Wall and do not seek to address the many other important aspects of this monument, including its construction, function and integration into the landscape, which will be explored in subsequent publications.
The Darband Wall
The remains of the large structure known as the Darband Wall, together with its colluvial debris accumulation, form a solid rampart that runs roughly north–south across the Shurob Valley between two mountain massifs belonging to the south-western spurs of the Hissar Mountains (Sarymas to the north and Susiztag to the south) (Figure 2). The rampart de facto enclosed this valley and thus allowed control of movement between the valleys of Kichik Ura Darya and Sherabad Darya/Machay Darya. As a result, it was likely an important checkpoint on the route between the present-day Qashqadaryo and Surxondaryo provinces, and hence the historic areas of Sogdiana and Bactria.
Figure 2. The Darband Wall as seen from the north (atop Sarymas Mountain) (photograph by L. Stančo).
The wall extends approximately 1.1km, starting from the steep slopes of Sarymas Mountain in the north and crossing the valley to the left bank of the Shurob river gorge in the south, where the slopes of Susiztag Mountain form the right bank (Figure 2). The location of the southern terminus is not as obvious as its northern counterpart. Despite erosion and the destruction of segments for the passage of an old road, a railway line and a new motorway, the wall is well preserved. Natural and artificial segmentation means that the wall currently stands in three sections: the northern (220m long), central (265m long) and southern (at least 335m long) sections. The wall does not form a straight line but breaks several times, taking advantage of changes in terrain, especially in elevation, and largely occupies the elevated ground of the small hills in the central part of the valley. The wall was reinforced with several towers, and part of the wall body and some towers have been archaeologically investigated.
Our fieldwork
The broader context of the Czech-Uzbekistani research project conducted in 2017–2019 is published elsewhere (Stančo et al. Reference Stančo2018, Reference Stančo, Shaydullaev, Khamidov, Augustinová, Damašek, Bek and Kmošek2019). Fieldwork was strategically designed to minimise destructive intervention to the already significantly damaged monument by focusing on an existing cut through the wall (see description below). For the sake of clarity, we designated this trench 10, resuming the earlier numbering of the French-Uzbek team (see Figure 3). The cut was cleaned over two days (23 and 27 September 2019), then stratigraphic units were described, key features measured, and photographs taken by drone and by hand for photogrammetric modelling of the section. On 29 September 2019 a series of charcoal samples were secured for radiocarbon dating.
Figure 3. Topographic plan of the entire Darband Wall with numbered locations of the earlier trenches (figure by Christian Mayer after Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 240, fig. 7).
Description of the Darband Wall section
The positioning of trench 10 was possible due to the extensive earthworks undertaken in the construction of the Samarkand-Termez highway. Although the highway was built on the route of the original 1980s road (part of the so-called Great Uzbek tract/highway), its construction required a substantial widening of the entire corridor and satellite images record disturbance to the wall between 2007 and 2019 (Figure 4). Trench 10 occupies the third and highest level of narrow artificial terraces dividing the steep slope that overlooks the highway. Destruction of this part of the wall is assumed to derive directly from bulldozer activity. The section (given the co-ordinates 38°12′36.8496″N, 66°58′22.0476″E) is south of the highway and is north facing. As a result, the section is protected from the sun and does not dry out too quickly, making it suitable for documentation. The overall width of the section reaches almost 20m, while the documented part is approximately 17m wide. The preserved height of the wall and its substructures reach around 5m at this particular spot, and within this we were able to document 22 stratigraphic units (Figure 5, Table 1).
Figure 4. Satellite images showing the locations of archaeological excavations, including stratigraphic sections from 1997 (trench 2) and 2019 (trench 10), only a few metres apart. The images compare the situation in 2007 (above) and 2022 (below). Substantial damage to the monument in the intervening period is apparent (based on images from Google Earth Pro compiled by L. Stančo).
Figure 5. The stratigraphic section of trench 10. Above) photograph taken by drone; below) photogrammetry overlaid with the numbers of stratigraphic units (see Table 1) (figure by T. Votroubeková).
Table 1. Stratigraphic units documented in the excavated section of the Darband Wall (trench 10).
Comparison to the earlier sections
The first documented section through the wall was created under similar circumstances; while the earlier road was constructed in the 1980s, E.V. Rtveladze recorded a stratigraphic section in a cut made by bulldozers (trench 1; Rtveladze Reference Rtveladze1986: 38, fig. 3). Rtveladze assigned an age to the wall based on the size of the mudbricks used for the outer walls (320–330 × 320–330 × 110mm), which matched those encountered at several sites dating to the Kushan Period throughout the Surxondaryo Province, and “the same dating is confirmed by fragments of pottery found in the wall, in particular the lower half of a cylindrical cup on a ring base, typical for that period” (Rtveladze Reference Rtveladze1986: 38). In absolute terms, both the first–third and the first–second centuries AD are suggested (Rtveladze Reference Rtveladze1986: 38–39). Unfortunately, neither the cup fragment nor any other material from this section was actually published to verify this age.
The second section of the wall was excavated in 1989/90 by the ‘Darband archaeology detachment’ of the Institute of Archaeology in Samarqand. The location of this section was approximately 10m—if not less—to the north of our later section (see Figure 4, trench 2). The section was not finished until 1997 (by an Uzbek-French team) but four phases were established (Rakhmanov Reference Rakhmanov and Masson1994). Dating of the individual phases of the wall was based upon the size of the mudbricks (420–430 × 100–120mm), a few pottery sherds (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 242, fig. 8) and two coins. Although descriptions of the pottery fragments were published (Rakhmanov Reference Rakhmanov and Masson1994: 76–77), these were not accompanied by photographs or drawings. An assemblage of pottery from Old Termez—located approximately 5km north-west of modern Termez, the administrative capital of Surxondaryo Province—was used as a reference collection for dating, and much of this collection dated to the Greco-Bactrian period. Rakhmanov (Reference Rakhmanov and Masson1994: 78) mentions that even earlier pottery fragments (mid-first millennium BC) were recognised among the surface finds, but again no pictures are provided. The two coins were attributed as imitations, one depicting the last Greco-Bactrian king Heliocles (end of the second century to first century BC) and the other similar to so-called Soter Megas coins belonging to the Kushan ruler Vima Takto (end of the first century AD). A 3.6m-wide wall of mudbricks with an internal corridor was identified on the west side of the section (marked with the letter H in Figure 6) and is interpreted here as a Hellenistic phase (meaning in this case the Greco-Bactrian period, third–second centuries BC) of the fortification, as excavations in trench 3 revealed a similar alleged Hellenistic wall that is 2.75m wide (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 241). The stone wall in the middle of the section is interpreted as a phase of fortification dating from the Kushan period.
Figure 6. The stratigraphic section of the Darband Wall excavated in 1997 (trench 2) (after Rapin Reference Rapin, Lindström, Hansen, Wieczorek and Tellenbach2013: fig. 10).
Charcoal sampling and processing
Samples of organic matter were collected from the section, selectively on 27 September 2019 (contexts DW-SS 002 and 005) and systematically on 29 September 2019 after its complete cleaning. This careful cleaning, during which approximately 50–100mm of the vertical surface of the cut was removed, was intended to ensure the purity of the samples. Sampling targeted larger charcoal fragments, presumed to be the charred remains of young trees or tree branches between two and 16 years old, allowing us to exclude the possibility of the so-called ‘old-wood effect’ error in our results. The charcoal was sampled from cultural layers and destruction horizons (including contexts DW-SS 001, 004, 005, 009, 010, 013 and 020), none of which were directly part of the masonry (mudbricks), reducing the likelihood of an older sample having intruded into a younger context.
Anthracological analysis was performed by Maria Hajnalová at the Department of Archaeology at Constantine the Philosopher University in Nitra, Slovakia, and a total of 10 samples derived from short-lived species were submitted for radiocarbon dating (Table 2). Samples were divided into two chronological phases based on context (Figure 7). One of the samples was obtained during preliminary cleaning of the profile, the other nine were sampled subsequently during documentation of the section. The selected samples were submitted to the Czech Radiocarbon Laboratory (Nuclear Physics Institute, Czech Academy of Sciences) for age determination using accelerator mass spectrometry. All samples represent carbonised plant remains.
Table 2. Results of radiocarbon dating on charcoal samples.
Dates were calibrated using OxCal v4.4.4 (Bronk Ramsey Reference Bronk Ramsey2021) and the IntCal20 calibration curve (Reimer et al. Reference Reimer2020).
Figure 7. The Bayesian distribution and summed probability of the calibrated radiocarbon dates for the Darband Wall, modelled using OxCal v.4.4.4 (figure by authors).
Radiocarbon dating: interpretation
Although the research presented here makes use of only a small part of the available archaeological data from the Darband Wall, some preliminary interpretations can be made with the expectation that they can be further refined in the future. The first concerns the timing of the monument’s foundation. Bayesian modelling of the radiocarbon dates indicates that the Darband Wall was probably built in the third century BC, most likely around mid-century (Figure 7), although the error margin covers several decades. Construction of the wall would therefore be associated with the period of Seleucid rule and the consolidation of new Asian territories, or perhaps with the early Greco-Bactrian rulers, the Diodotids, at the latest. The foundation of the wall at the time of Alexander’s Asian campaign (329–327 BC) seems to be highly unlikely.
Six of the 10 analysed samples provide very close absolute dates in the range 390/359–200/179 BC, with a high degree of probability that these dates fall within the third century BC. If placed directly in the stratigraphic section, into this phase falls: 1) the pakhsa (compacted clay forming solid walls in large blocks or layers) platform/substructure for the main wall (stratigraphic unit DW-SS 009; Figure 5); 2) the layer below this pakhsa platform (three samples from DW-SS 013); 3) the main stone-made wall itself (DW-SS 010); and 4) destruction layers to the west of it (DW-SS 020). This latter inclusion seems to indicate that these layers were originally part of, or related to, the structure before erosion began.
The rest of the dated samples relate to a phase of reconstruction and reinforcement of the wall. These activities took place at the beginning of the first millennium AD, although closer inspection indicates that there are actually two pairs of dates. The first two relate to the first decades of the first century AD (range 48/40 BC–AD 108/80), the second to a period 60–80 years later, at the turn of the first and second centuries AD. The earlier pair of dates derive from a destruction/erosive layer (DW-SS 001) later in date than the white gypsum fill of the ditch (DW-SS 015). The ditch and its fill are two distinct events in time; we suggest that the original ditch was contemporary with the stone wall, while the fill relates to the levelling prior to the later phase (mudbrick wall).
Based on available stratigraphic and radiometric data, we therefore hypothesise that events unfolded as follows. In the first half/middle of the third century BC the gypsum subsoil was levelled to allow construction of a pakhsa platform (DW-SS 009) and the 1.3m-thick stone wall (original height unknown) (DW-SS 007). Shortly thereafter, the larger stone wall (DW-SS 010) was built on top of the platform and a ditch was dug into the subsoil to the east of the walls. The depth of the ditch is not currently known, since our section did not reach the subsoil in its easternmost part. This arrangement is reminiscent of early Greek fortifications in the Classical period, presenting a combination of a main wall (teichos) with a lower fore wall (proteichisma), preventing the enemy from reaching the main one, and a ditch (taphros). We draw this comparison tentatively, as we do not have a date for the smaller wall (proteichisma), only for the layer on which both walls were built. However, the smaller wall cannot be earlier than the larger wall, as the position of the smaller wall, apparently in lower place, might suggest. The relatively small distance between the main wall and the supposed proteichisma—just 2m—is also problematic. Relatively small, yet distinctly different, distances between the main wall and the proteichismata have been identified in Hellenic fortifications in Central Asia: at Karabag Tepa (Khalchayan) the distance between these elements was said to be only 0.8m, while at Keikobad Shakh this distance reached 3.5m and at Khazarasp (Khwarezm) there is the distance of 13.5m (Pugachenkova Reference Pugachenkova1966: 125). The fore wall became a common feature of Greek fortifications in the Hellenistic period at the latest; the Athenian Walls (1.1–2.5m in width), the most famous example, were constructed in the fourth century BC, while the adjacent moat is dated to the second half of the same century (Parigi Reference Parigi, Frederiksen, Müth, Schneider and Schnelle2016: 385). Perhaps coincidently, the Athenian moat was filled and the proteichisma went out of use also in the first century BC to first century AD (Parigi Reference Parigi, Frederiksen, Müth, Schneider and Schnelle2016: 393).
From the available evidence, the stone-made wall in the lower part of the valley does not seem to have had a counterpart on the central ridge to the south (location of trench 3, see Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 241–42), where only the remains of a mudbrick wall, dated by the excavators to the third century BC but probably younger, have been documented. It is therefore possible that the original third-century BC fortification enclosed only the most vulnerable parts of the valley.
Regardless, reading the situation as a larger main wall with a shorter fore wall and fronting ditch would necessitate a reorientation of the wall that fundamentally alters its interpretation. The original phase of the wall in this scenario is oriented towards the east with the defenders/builders being on the west side. Almost without exception, this is the opposite of what has previously been considered and would have a dramatic impact on our understanding of historical events that took place in the Bactro-Sogdian border region (see below). Further support for this reorientation can be found in trench 3, where part of a tower was excavated on the eastern side of the rampart. The excavators interpret this as an inner tower (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 241), but its presence makes more sense if we reverse our concept of the direction in which the wall is facing, whereby the tower assumes a more standard formation outside the wall, preventing direct attacks on its body.
Although we do not have a date for the intentional filling of the ditch (DW-SS 015), the dating of the layer above this event (DW-SS 001) establishes the terminus ante quem precisely at the turn of the first century BC and the first century AD, or at the end of the first century—when the new mudbrick wall was built (see below)—at the latest. As the ditch no longer existed in this period, it is possible that the orientation of the fortification was reversed in the Kushan period. The fundamental reinforcement of the wall, at least in this section, then came at the end of the first century AD or during the first half of the second century, as shown by the two dates from the foundation of the mudbrick wall reinforcement (DW-SS 005, 004).
Due to the lack of written sources, we can only speculate as to the reasons for strengthening the wall perhaps even before the accession of the Kushan king Kanishka. The change in the orientation of the wall itself can be interpreted as an effort to establish a new border for the emerging Kushan empire or to strengthen an existing border, perhaps in an effort to secure the northern/north-facing border of the Kushan domains at a time of peak expansion to the south-east, into India. At this time, the builders/defenders of the wall were now situated to the east and south of the wall, while the west and north were outside and posed some threat. The Kangju confederation based in Sogdiana represents the most obvious opponent here. We emphasise, however, that this putative changing of orientation requires further corroboration from both the archaeological and historical evidence, and so must remain a hypothesis at present.
Discussion
The dating of historic buildings depends on a number of factors, in the case of the Darband Wall these include the large scale of the structure and the apparent individual development of each of its parts. Even after the results of the Uzbek-French expedition, including the new dating, were published, Rtveladze insisted—without providing any new evidence—that the fortification was Early Kushan in date and that no wall stood in the area of trench 1 in the Greco-Bactrian period (Rtveladze Reference Rtveladze2002: 115). The stratigraphic sections from trenches 1, 2 and 10 show a quite different building situation and history, and any stratigraphic and chronological comparison is thus hampered.
Previous attempts to date other parts of the structure using such indicators as the size of mudbricks or individual fragments of ceramic vessels are considered inconclusive, although this type of data must be used to inform estimates. Yet, for ceramics in particular, we would expect a large, statistically relevant body of pottery to be evaluated, originating optimally from closed contexts. The size of the mudbricks (420–430 × 100mm), which has been used as an indicator of a Hellenistic date (Rapin et al. Reference Rapin, Khasanov, Rakhmanov, Baumer, Novák and Rutishauser2022: 241), also corresponds, for example, to bricks found at Dal´verzintepa and Jandavlattepa dated to the Kushan (first–third centuries AD) and Kushan-Sasanian (third–fourth centuries AD) period (Pugachenkova Reference Pugachenkova, Pugachenkova and Rtveladze1978: 29; Stančo Reference Stančo, Abdullaev and Stančo2011: 36), and as such cannot be used as a primary chronological indicator. On the other hand, comparison with the masonry used in the construction of the Uzundara fortress (see Figure 1) can be considered a good argument for confirming the dating of the stone wall at Darband to the Hellenistic period. The Uzundara fortress was exclusively defensive architecture built of unworked stones; no part of the fort was built of mudbricks of any size (Dvurechenskaya Reference Dvurechenskaya, Baumer and Novák2019). The dating of the Uzundara foundation by excavators to no later than the reign of the Seleucid king Antiochus I (281–261 BC) therefore further supports our conclusions regarding the chronological position of the fortress system on the border of Bactria and Sogdiana (Dvurechenskaya Reference Dvurechenskaya, Baumer and Novák2019: 158).
A similar debate concerning the dating of fortifications and their attribution to a Greek or non-Greek builder also surrounds the city walls of Dura Europos in present day Syria. This originally Seleucid fortress on the Euphrates River, dating from the turn of the fourth and third centuries BC, had walls made partly of mudbrick and partly of stone. Previously, the mudbrick parts of the structure were considered to be original and therefore Greek (von Gerkan Reference von Gerkan, Rostovtzeff, Brown and Welles1936), while the stone parts were thought not to have been added until the Parthian rebuilding (113 BC–AD 165). A more recent interpretation (Leriche Reference Leriche2010: 29–31) suggests Greek walls of ashlar stone, started in the mid-second century BC, were hastily completed a few decades later with unburnt brick during the Parthian threat—that is, before the Parthians took control of the city and rebuilt its fortifications. The later phase of Darband Wall reconstruction with mudbricks did not stand alone, but reinforced the existing stone wall, as happened on a larger scale in Dura Europos during the Sasanian invasion in the third century AD.
Our proposed dating of the early phase of the Darband Wall also finds indirect support from coins identified in the extensive surface collections of the Czech-Uzbekistani expeditions carried out in 2018, 2019 and, most recently, in 2022. The first two seasons recovered hundreds of coins, including 61 of demonstrably Hellenistic vintage (i.e. dated from the third–second centuries BC) and an overall range that covers 12 rulers, starting with the so-called Alexander type (c. 300 BC), through Antiochus I, Diodotus I and II, Antiochus of Bactria, Euthydemus I, Demetrius I, Antimachus, Pantaleon and Agathocles, to Eucratides I and Heliocles in the third quarter of the second century BC (Stančo Reference Stančo2021: 81–82; Stančo et al. Reference Stančo, Shaydullaev, Militký, Kmošek and Bek2024). Half of these coins (31 specimens) were minted under Euthydemus I, a situation that is also found at the nearby fortress of Uzundara (Gorin & Dvurechenskaya Reference Gorin and Dvurechenskaya2018; Stančo Reference Stančo2021: 81–82). This high frequency and the even distribution of finds across the site indicate intensive use of the fortification during the last third of the third century BC. Ten coins, including a silver drachma of the ‘Alexander type’ minted around the year 300 BC, are, however, earlier in date and could in themselves lead to the assumption that the wall dates back to the period of Seleucid rule (to Diodotus at the latest, i.e. c. 255–235 BC). Yet, none of the building phases dated here fall clearly into the period of the end of the third or the beginning of the second century BC. Although the use of this area may have been intensive at that time, there do not appear to have been any substantial contemporaneous reconstruction or strengthening efforts in the section we examined.
Conclusion
The Darband Wall is a historical and archaeological monument of regional and global significance that has been the subject of intense research interest in the past 40 years. Accurate dating and interpretation of the wall can help answer important questions about the political and economic history of this part of Central Asia, as well as deepening our understanding of the mobility of populations in both the early and later medieval periods. Here, we have sought to take a first step towards the refinement of the chronology of this fortification, and to take a second step in offering a different perspective on the subsequent interpretation of this monument.
Based on the radiocarbon dating of 10 charcoal samples taken from a section of the wall, the first—and perhaps also the most structurally significant—phase of fortification at Darband was constructed in the first half/middle of the third century BC, half a century earlier than previously estimated (Stančo Reference Stančo2021). Thus, during the establishment of the Seleucid state in the territory of Bactria and Sogdiana—or during the rule of the first independent Greco-Bactrian king Diodotus—an extensive wall was built on the border between these territorial units. The early date suggests that the wall did not arise primarily from a need to isolate the settled ‘Greek’ Bactria from the nomads threatening it from the north in the late third century BC, though it may have served this purpose later on. Ultimately, the wall protected political entity A in the north-west (Sogdiana) from entity B in the south-east (Bactria), but the question of who these entities represented in the first half of the third century BC remains to be answered.
Funding statement
This work was supported by the European Regional Development Fund project ‘Creativity and adaptability as conditions of the success of Europe in an interrelated world’ (reg. no.: CZ.02.1.01/0.0/0.0/16_019/0000734). Publication was supported by the Cooperation Program provided by Charles University, research area Arch, implemented at the Faculty of Arts of Charles University. The research is part of the broader topic studied within the ‘Ready for the future: understanding long-term resilience of the human culture’ project (CZ.02.01.01/00/22_008/0004593).
Data availability statement
The authors confirm that the data supporting the findings of this study are available within the article.
Author Contribution: CRediT Taxonomy
Ladislav Stančo: Conceptualization-Lead, Funding acquisition-Lead, Methodology-Equal, Project administration-Lead, Resources-Lead, Supervision-Lead, Writing - original draft-Lead, Writing - review & editing-Lead. Jan Petřík: Conceptualization-Supporting, Data curation-Equal, Funding acquisition-Supporting, Investigation-Equal, Methodology-Equal, Validation-Equal, Writing - original draft-Equal, Writing - review & editing-Equal. Mária Hajnalová: Data curation-Equal, Formal analysis-Equal, Investigation-Equal, Writing - original draft-Supporting, Writing - review & editing-Supporting. Tatiana Votroubeková: Data curation-Supporting, Formal analysis-Supporting, Investigation-Equal, Writing - original draft-Supporting, Writing - review & editing-Supporting. Shapulat Shaydullaev: Conceptualization-Supporting, Resources-Supporting, Supervision-Equal, Validation-Supporting.
Open access
