The medieval croft plužina field system in a mountain region of central Europe: The interdisciplinary record of the earthen field boundaries in Debrné, Czechia

The integration of archaeological, historical and geoarchaeological records represents a significant contribution to research into the medieval landscape. This study focuses on the medieval field system in the deserted village of Debrné, located in northeastern Bohemia, Czechia. The village features a well‐preserved croft plužina field system, a typical historical landscape of central Europe. The main and most visible elements of the field system are the earthen field boundaries, which were the focus of the geoarchaeological investigations. Archaeological excavations in trench S1 revealed a collection of larger stones at a depth of 1 m, potentially serving a drainage function akin to the observed plužina. Additionally, a boulder paving, identified as a remnant of a path between fields, provided insights into the historical use of the area. In trench S2, positioned closer to the village's core, layers with increasing stone content were recorded at a depth of 130 cm. However, the drainage structure observed in trench S1 was not replicated. The dating of earthen field boundaries indicated the creation of the terrace in the second half of the 16th century in trench S1. In trench S2, radiocarbon dating at a depth of 70 to 80 cm ranged from 1025 to 1175 A.D., predating the first written source about Debrné. Optically stimulated luminescence (OSL) dating in trench S2 suggested exposure to sunlight during the third century A.D. Archaeobotanical analysis of charred macroremains from trench S1 revealed 236 plant macroremains belonging to approximately 20 taxa, showcasing wild‐growing, collected useful species such as raspberry, blackberry and elderberry. Trench S2 yielded 23 plant macroremains belonging to 11 taxa, with similar species as trench S1. Pedological and micromorphological examinations displayed distinct layering in both trenches, showing up to six layers. Micromorphological analysis unveiled vuggy microstructures, varying grain sizes and elemental patterns, shedding light on different periods of occupation. Multidisciplinary investigations of the buried soils forming the earthen field boundaries discovered that the original soil cover comprised automorphic lixisols and cambisols, which also form under present conditions. These results underscore the importance of integrating pedological, geoarchaeological, archaeobotanical and physical data to comprehend the intricate nature of anthropogenic landscape changes.

DG18P02OVV060; IGA, Faculty of Environmental Sciences, CZU Prague, Grant/Award Numbers: 2021B0046, B2020B0042 earthen field boundaries discovered that the original soil cover comprised automorphic lixisols and cambisols, which also form under present conditions.These results underscore the importance of integrating pedological, geoarchaeological, archaeobotanical and physical data to comprehend the intricate nature of anthropogenic landscape changes.

| INTRODUCTION
Research on historical field systems is developing rapidly (Brown et al., 2020(Brown et al., , 2021(Brown et al., , 2023;;Kinnaird et al., 2021;Turner et al., 2021;Vervust et al., 2020).The genesis of different types of landscapes in Europe (Lebeau, 1996;Poschlod, 2015), as well as the historical and ecological factors of parcel boundaries (Kinnaird et al., 2021;Turner et al., 2021) as the main features of field systems in mountainous and foothill landscapes are at the forefront of interest (Fanta et al., 2022;Keller et al., 2023;Silva-Pérez & González-Romero, 2022;Vervust et al., 2020).In the field of research on agricultural settlements in central Europe, the study of the typology and chronology of earthen field-boundary structures (a variation to agricultural terraces) as a key component of the historical landscape has developed (Beneš et al., 2022;Čapek & Holata, 2017;Fanta et al., 2020Fanta et al., , 2022;;Houfková et al., 2015;Poledník Mohammadi et al., 2023;Shetti et al., 2022;Šitnerová et al., 2020;Škabrada, 2022).Much analysis of landscape components focuses on the spatial organization and physical structure of such a field system (plužina in the Czech language, der Flur in German), which is understood as the sum of all agricultural areas of one village or other historical settlement (hamlet, small town, castle, etc.)In general, it is characterized as an economically usable part of the landscape belonging to a village settlement; it is a summary of all fields, meadows and pastures connected to the settlement by a network of roads or paths (Beneš et al., 2022;Gojda, 2000;Štěpánek, 1967).Sometimes, only the cultivated (ploughed) area is understood as the agricultural part of the estate's grounds divided into individual parcels, and meadows, pastures and forests are not included (Dohnal, 2003).Moreover, some authors use the term plužina only for the physical remnants of historical landscape structures (e.g., Čulík et al., 2015).The plužina and its variants represent a typical agricultural landscape setting in central Europe.The basic unit of the plužina is the field parcel or any particular field.The most characteristic sign is a hedgerow that separates single fields into the shape of land parcels.These hedgerows are usually made of stony walls or layered earth commonly overgrown by trees and shrubs.Hedgerows are essential for a well-functioning landscape because they reduce soil erosion and retain water (Bayer & Beneš, 2004;Burel, 1996;Ouin & Burel, 2002).In landscape ecology, this feature is regarded as an important landscape unit essential for the ecological stability of the agricultural landscape and as an important historical and aesthetic attribute that is going to be accepted as a part of cultural heritage (Fanta et al., 2022;Kučová, 2018;Nitra, 2018;Zacharová et al., 2022).
The terraced and open field systems are characterized by distinct boundaries among the parcels, among which the plužina field system can be classified into this category.While the plužina field system shares similarities with lynchets, there are notable differences that preclude its classification as such.This distinction holds despite the study by Jackovičová et al. (2023), who used lynchets as a synonym for the entire plužina system.
First, the formation process differs.Lynchets tend to be created by ploughing strips in one direction (Curwen, 1939), while the plužina field system was deliberately created during the foundation of the adjacent village (Šitnerová et al., 2020).This is why the plužina around a settlement represents a unique landscape type, one that has been deliberately constructed by humans.
Second, most types of the plužina are related to the linear villages in the valley.In these cases, the parcels of each homestead were quite hilly, and the fields and their boundaries were laid parallel next to each other and copied the gradient curve (Figure 1), so the fields could not create terraces.
The field system in Czechia could have been organized in several different ways.We have descriptions of six (Löw & Míchal, 2003) to nine types (Černý, 1973, 1979)  or historical processes (Fanta et al., 2022).The segmental plužina is bound to places with poor soil fertility and very rugged terrain and also to villages founded during the early medieval period.The sectional plužina is primarily found in regions with perfect conditions for agriculture (flat terrain, very high soil fertility), and often where colonization occurred during the high medieval period.The croft plužina (the case of Debrné), mainly connected to villages with regular hides, is characteristic of late medieval colonization (Figure 1).Furthermore, while some plužina types are regularly distributed across Czechia, others occur only in specific regions (Fanta et al., 2022).A different typology of the plužina, reflecting the social and economic relationships within the village society, has been recently proposed by Klír (2020).The Debrné field system study provided a well-preserved archaeological record of soils and sediments dating from the high medieval period to the early modern era.
The main aim of this study is to investigate the construction of the earthen field boundaries of agricultural field parcels and obtain material for dating their origin and further development and pedogenesis under the influence of human activity (Beneš et al., 2022;Houfková et al., 2015;Šitnerová et al., 2020).In particular, geoarchaeological sampling is aimed at gathering information about the area to understand how the geochemical and sedimentological compositions of the archaeological soils vary over space and time.
Finally, we want to find out to what extent, in the case of croft plužina, the people of the Middle Ages were engaged in bringing about a fundamental and persistent change in the landscape.

| Geology and climatic factors
The cadastral area of the abandoned village of Debrné (Döberle in German) (N 50°36′ 09.847″ E 015°58′ 36.406″)(Figure 2) is situated in the foothills of the Krkonoše ('Giant') Mountains near the district town of Trutnov.The altitude of the study site ranges from 450 to 679 masl.The territory of Debrné is part of the Inner Sudetes Basin.The bedrock mainly comprises siltstones, sandstones and, locally, red-brown variegated upper carboniferous siltstones and sandstones (https://www.geology.cz).Locally, bands of volcanic origin-tuffs and andesites-occur.Intense weathering under tropical climate conditions produced an Fe-rich geological substrate.At higher elevations, cambisols with isolated areas of pseudogley soils predominate (https://www.vumop.cz).Climatically, the area falls on the borderline between moderately warm and cool areas with relatively higher rainfall (Quitt, 1971).The potential natural F I G U R E 1 Spatial distribution of the croft plužina within Bohemia (the western part of the Czech Republic); the percentage scale shows the share of croft plužina type among all other plužina types.A small red cross marks the position of Debrné.Source: Fanta et al. (2022).The black-and-white schemes represent the spatial structure of fields in specific plužina types according to Černý's typology (black-and-white illustrations taken from Černý, 1979).vegetation map represents a beech forest over the wider area, and a beech or fir oak forest at higher elevations (Neuhäuslová et al., 2001).

| Archaeology of the region and the site of Debrné
The Trutnov region has a long history of human occupation, with evidence of hunter-gatherers dating back to the Late Paleolithic and Mesolithic periods.Agricultural sites from the Neolithic period onwards were mainly found in the lower altitudes of the district, while the higher altitudes surrounding Trutnov remained unsettled until the Middle Ages.The establishment of a trade route to Silesia was the reason why the landscape became intensively occupied.
Notably, the oldest settlements in the Krkonoše ('Giant') Mountains are documented in this region (Honl, 1967;Wolf, 2000).One of the first settled villages was Debrné.The first written sources date to 1260 A.D., when the village of 'Debrné u Trutnova' was donated by Idik of Úpa to the Trutnov nursing home.The hospital was administered by the Crusaders of Zderaz, who later leased the village to the town of Trutnov and sold it in 1580 (Hrabová & Beneš, 1967).At the beginning of the 19th century, an exclusively German-speaking population is mentioned in Debrné; altogether, 73 houses with almost 456 inhabitants are indicated here (Beneš et al., 2022).Nearly all the inhabitants were displaced after the Second World War and new arrivals did not want to live in a remote village in the mountains.The number of inhabitants gradually decreased, and the houses decayed till the 1950s, when a thermal power station was built nearby.Part of the village was flooded because of the building of the reservoir 'Dead Lake' (Mrtvé jezero), which served as an ash disposal site.The rest of the village was cut off from the road due to the flooding, the village finally declining in the 1960s.Terrain relics of the vanished buildings are still visible, as are several crosses and the ruins of the chapel of St.
John the Baptist (Figure 3).Within the framework of the larger research field system project covering all of Czechia, attention was paid to the identification, diversity and archaeological parameters of the preserved remains of the historical plužina.Five field systems, representing their different forms in different landscape types, were selected for the environmental archaeology investigation (Beneš et al., 2022;Šitnerová et al., 2020).The purpose of the overall project, which also included the research on the field systems in Debrné, is to date the parcel boundaries accurately, and obtain data that would allow a more detailed knowledge, description and classification of this landscape phenomenon.

| Archaeological excavation
Most of the environmental data were sampled from two field trenches.Their location was determined based on the distribution, spanning from a point near the core of the abandoned village (S2) to a point distant from the settlement (S1).While trench S1 provided material for a limited set of analytical methods, we concentrated the broader effort on trench S2 to obtain the most detailed transdisciplinary evidence (Figure 4).
Both archaeological trenches, S1 (N 50°36′ 33″ E 015°58′ 23″) and S2 (N 50°36′ 08″ E 015°58′ 16″) were within the field system of the abandoned village Debrné (Figure 5).Locations were selected at different distances from the abandoned village to capture the possible differences in parcel boundary characteristics.The first trench S1 (Figure 4a), measuring 5 × 1 m, was placed closer to the edge of the preserved terrace at a distance of 750 m from the centre of the village (Figure 3).The terrain was surveyed from the surface in 10 cm physical layers with photographic documentation.A soil sample was collected from each level.The depth of the trench reached 100 cm.The second trench S2 (6 × 1 m) (Figure 4b) was placed 200 m from the core of the abandoned village (Figure 2).Within S2, we recorded compact layers with gradually increasing amounts of stones.The depth of the trench reached 130 cm.Trench S2 was also subjected to a detailed pedological and micromorphological investigation and magnetic susceptibility measurements.From a depth of 70 cm, a sample was taken for dating by optically stimulated luminescence (OSL).Samples were then collected from the trench for environmental analyses, and analysis of the radionuclides 210 Pb and 137 Cs (Houfková et al., 2015).

| Archaeobotany
Soil samples were collected for plant macroremains analysis during the digging of the trenches.Samples were collected in mechanical layers of 10 cm each, as soil horizons are usually not well discernible visually during the digging.Samples (10 L each) were extracted by water flotation, using a flotation tank of ANAKARA type.The light fraction was collected on sieves with a mesh size of 0.25 mm.Uncharred and charred remains were used for this study (Anderberg, 1994;Berggren, 1981).Botanical macroremains (Supporting Information S1: Table S2) were identified using a standard binocular microscope Nikon C-LEDS with ×10-×50 magnification.

| Sampling protocols
The sedimentary archive at the Debrné site manifests pronounced variability attributable to the impact of the topographic slope on the deposition of soil constituents.Describing the soil samples involved evaluating and documenting various soil layers or horizon characteristics.After digging the trenches as a vertical cut into the soil, the depth and size of the trenches were based on our field objectives and the position of the bedrock or geological substrate.
The soil horizons were identified using various properties such as colour, changes of material, texture, structure consistency, moisture content, organic matter, roots and fauna.Depths of each horizon were recorded from the soil surface to the bedrock.Forty-six soil samples from each 5 cm of profile for further analysis such as laboratory testing were collected.Five thin sections measuring 4 × 7 cm in diameter were sampled for micromorphology.The micromorphological samples were taken from each horizon where any lithological change was visible, then packed into cling film and transported to the laboratory.After being dried and impregnated in a vacuum by resin Polylite 2000, thin sections with a thickness of 30 µm were made from the samples.These sections (Figures 6 and 7) were observed under a polarizing microscope at magnifications ranging from ×16-×400 and described according to Stoops (2003) and Stoops et al. (2010).
Sediments were identified in their wet states using a Munsell soil colour chart in the field (Munsell, 1905) along with other factors such as texture, structure and consistency, which were used to determine and specify soil layers and group soils based on a standard soil classification system (FAO, 2021).

| Radiocarbon and OSL dating
One of the main dating methods is radiocarbon AMS 14 C, utilizing charred seeds from annual plants and charcoal as suitable dating materials.Trench S1 yielded dates for three levels: the depth of 60-70 cm was dated using macroremains of charred Rubus seeds, | 433 while levels 80-90 and 90-100 cm were dated using charcoal.In trench S2, two levels were dated using AMS 14 C (Figures 8 and 9): the level at 30-40 cm was dated using charred caryopsis of Pisum/Vicia seeds and the level at 70-80 cm was dated using charred macroremains of Rubus seeds (Supporting Information S1: Table S2).All samples were subjected to analysis at the Poznan Radiocarbon Laboratory in Poland and were calibrated with OxCal based on the IntCal13 atmospheric curve (Bronk-Ramsey, 2009;Reimer et al., 2013).For enhanced data reliability, OSL was used to date the level at 70 cm in trench S2.This sample was processed at the Gliwice Absolute Dating Methods Centre in Poland (Moska et al., 2021) and analysed using the single-aliquot-regeneration method (Bøtter-Jensen et al., 2010;Wintle & Murray, 2006).Quartz grains (45-63 μm) were extracted from sediment samples for OSL measurements.The OSL measurement was conducted using an automated Risø TL/OSL DA-20.The dose-response curves for SAR were most accurately modelled using a single saturating exponential function.
Final equivalent dose (D e ) values were calculated using the Central Age Model (CAM) (Galbraith et al., 1999) using the R package 'Luminescence' (Kreutzer et al., 2012(Kreutzer et al., , 2020)).This calculation involved an analysis of dose distribution (Figure 9; Berger, 2010) and consideration of an overdispersion parameter, which was determined to be 10% for the analysed sample.The presence of a unimodal dose distribution and a low overdispersion parameter value indicated that the tested material represented a well-bleached quartz group (Moska, 2019), supporting the application of the CAM model for final equivalent dose calculations.High-resolution Canberra gamma spectrometry was utilized to evaluate dose rates arising from decay chains and potassium.The spectrometry measurements were calibrated using reference materials, including IAEA-RGU-1, IAEA-RGTh-1 and IAEA-RGK-1 from the International Atomic Energy Agency.Dose rates were computed through an on-line dose rate calculator (Tudyka et al., 2023) that incorporates the latest conversion factors.It was assumed that the average water content did not exceed | 435 15%, leading to the use of a value of 15 ± 3% for subsequent calculations (Supporting Information S1: Table S1).

| Measurement of short-lived isotopes to exclude major soil disturbances
Determination of the content of short-lived isotopes is important to exclude major soil disturbances and to check the integrity of the stratigraphy.It was used for the purposes of recent soil dating as a method that uses radioactive isotopes of lead ( 210 Pb) and caesium ( 137 Cs) to detect stirred sediments.Samples were collected at each 5 cm from the profile of trench S2.About 7 g of sediments were dried by lyophilization and measured in the CEN Radiochronology Laboratory in Canada using a high-purity germanium detector (Figure 10).

| Elemental analysis
To prepare the samples for portable X-ray fluorescence (pXRF) (Kalnicky & Singhvi, 2001;Markowicz & Van Grieken, 2002;Sitko, 2009) analysis, they were first dried at 40°C for 24 h, then sieved through a 2-mm mesh F I G U R E 7 Micromorphological documentation of the main differences between soils accumulated in different timespans at Debrné, trench S2: (a) Angular blocky aggregates: at this low magnification, it is possible to see the general shape of the angular large peds with the presence of microcharcoals at 25 cm depth (PPL).(b) Crumb aggregates with a weak degree of separation with the presence of Fe/Mn nodules and microcharcoal at 67 cm depth (PPL).(c) Geodic nodule: Nodule showing an empty core, that is, an irregular-shaped void in a silty groundmass at 47 cm depth (PPL).(d) Unsorted uppermost soil horizon: There is a high appearance of daub and rock fragments in the locally depleted matrix with common passage features and Fe/Mn nodules at 67 cm depth (PPL).(e) Red and laminated clay coating at 110 cm depth (PPL).(F) Fissure/ crack (Bullock et al., 1985) microstructure: Blocky aggregates are not fully separated.Generally, the groundmass is dense, except for a few planes and possibly some channels, as shown in the picture with the presence of microcharcoal and Fe/Mn nodules at 67 cm depth (PPL).PPL, plane-polarized light.sieve and ground to the resulting fraction in a porcelain mortar.We used a portable ED-XRF analyser (Olympus InnovX 'Delta Professional') in Soil Geochem mode, measuring each sample for 30 s with a 10 kV beam (for better measurement of lighter elements) and for 30 s with a 40 kV beam (for better measurement of heavier elements).Results are presented in weight parts-per-million.The quality of the device measurements was successfully tested using 55 reference materials.This analytical process has been utilized in numerous previous studies (e.g., Horák et al., 2018;Janovský et al., 2020;Poledník Mohammadi et al., 2023) and is reliable and comparable to more established laboratory techniques (Save et al., 2020).
Statistical software R (version 3.6.0)was applied for the data analyses.The final data set (Supporting Information S1: Figures S1   and S2) consisted of 46 samples with the elements Al, Si, K, Rb, Sr, MN, Fe, Zn, As, Zr, Pb and LEs (light elements), and aggregate expression for the content of elements from H to Na, which can be detected by pXRF but cannot be distinguished.Although this is not an elemental content, but rather a quasielemental variable, we usually use it in the analyses anyway, as it bears proxy information (and usually tends to be connected to the topsoil or organic material) (Horák et al., 2018).The basic statistical characteristics are presented in Table 2, which is a suitable form for such tasks when working with geochemical data.The characteristics were mainly used for exploratory visualization techniques via boxplots and scatterplots.For testing the diversity between categories, we used the Kruskal-Wallis test.We also used multivariate analysis (principal component analysis [PCA] computed in R environment).Both approaches emphasize different aspects of the data set, and combining the two can reveal interesting patterns.

| Grain-size distribution and magnetic susceptibility
The grain-size distribution was measured at the Institute of Geology ASCR using a CILAS 1190 LD laser particle-size analyser (CILAS), providing a measurement range of 0.004-2.5 mm.Particle-size analysis was performed after short ultrasonic dispersion (60 s) on pretreated soil samples.Pretreatment of soil samples (fraction In trench S1, at a depth of 1 m, a number of larger stones were found. It is possible that these stones had a similar drainage function as the plužina observed at Valštejn (Šitnerová et al., 2020).At the edge of trench S1, a boulder paving was excavated, which was identified as a remnant of a path that ran through the hedgerow between fields (Figures 4 and 8).The road was in use until recently as there is still evidence of it on a topographic map from 1952; a piece of a motorcycle footplate was also discovered.Trench S2, located closer to the village's core, reached a depth of 130 cm.Brownish-black to red compact layers were recorded in both trenches.Gradually increasing amounts of stones were also recorded, but the drainage structure as in trench S1 was not captured (Figures 4 and 8).

| Archaeobotany
Charred plant macroremains were utilized to document historical vegetation.As is common in terrace field sediments, the concentration of plant macrofossils was extremely low.Even so, some shrubs were identified.The archaeobotanical work targeted charred macroremains, and analysed 10 samples from trench S1.Two hundred and thirty-six plant macroremains belonging to approximately 20 taxa were found (see the Supporting Information).Most of the finds were uncharred.No cultivated species were found.The charred finds mostly concerned wild-growing, collected useful species, such as raspberry (Rubus idaeus), blackberry (Rubus fruticosus) and black elderberry (Sambucus nigra) and another species of elderberry (Sambucus ebulus) (Table S2).
The archaeobotanical studies of seven samples from trench S2 Sambucus sp.) and strawberry (Fragaria vesca/viridis).The analysis results of the botanical charred and uncharred macroremains illustrate the state of vegetation in medieval times at the site of the archaeological trenches (Supporting Information S1: Table S2).

| Dating of the earthen field boundaries
The radiocarbon dates obtained from both trenches are summarized in Figure 11.For trench S1, the depth 60-70 cm was dated to the middle of the 17th century, the depth 80-90 cm to the turn of the 16th to 17th century and the subsoil beneath the drainage dates the creation of the terrace to the second half of the 16th century (Figure 11).Because S1 is in the more distant part of the plužina from the village, it may mean that these fields and their boundary strips were created during the additional 'field creation/landscaping' associated with the extension of the village.

| Measurement of short-lived isotopes to exclude major soil disturbances
The results of the measurement of radionuclide isotopes are illustrated in Figure 10.The concentrations of isotopes 210 Pb are extremely low, near the detection limit.The measurement results were negative and the emanation coefficient, usually 0.3, was used (Du & Walling, 2012).The radionuclide 137 Cs was present only in the upper part of the trench S2 profile.This radionuclide is anthropogenic with a half-life of 30 years, and it is associated with the use of nuclear weapons (Madsen et al., 2005).These findings are only in the first 30 cm, suggesting that the profile contains undisturbed soil.

| Pedology and micromorphology
The Debrné sedimentary record shows significant variation due to the influence of the site's slope on soil material accumulation (refer to Figures 4 and 8).At trench S1, macroscopically, up to six layers were distinguishable based on colour (field), and material, ranging from 2.5YR hues, and ranging from 3/2, 3/3 and 3/4 with textures of clay, sandyclay, and silty clay.Archaeologically, these layers have been linked to different occupational and geomorphological events during the 12th century (occupational events as determined by radiocarbon dating).
With the soil horizons studied at Debrné, trench S1 showed distinct morphological characteristics, including easily demarcated horizons with wavy, sharp or smooth topography.The A-horizon (Ah) was enriched with organic matter (Ah) and showed a dusky red material colour of 2.5YR 3/2.Above the Ah, we found the topsoil (AhB) to also show a dusky red material colour of 2.5YR 3/2.Moving down, the C1 horizon was characterized by a dark reddish-brown soil with a material colour of 2.5YR 3/3.This horizon contained charcoal and a substantial quantity of larger stones.Beneath the C1 horizon, the C2 horizon shared the same dark reddish-brown soil colour (2.5YR 3/3) but contained fewer stones.
Further down, the C3 horizon was identified by its dark reddish-brown soil with a material colour of 2.5YR 3/3 and the presence of larger stones.Finally, the R-horizon, which represented a defunct road, was F I G U R E 11 Radiocarbon data from trench S1 and S2.
| 439 also composed of dark reddish-brown soil with a material colour of 2.5YR 3/4 (Figure 8a).3/4) and contains smaller stones.Finally, at the base of the trench, there is the bedrock layer, characterized by a dark reddish-brown (2.5YR 3/4) colour.This layer includes larger stones and shows some presence of microcharcoals (Figure 8b).Micromorphological analysis of the Debrné samples (Table 1) from the S2 profile indicated fundamental differences between the archaeological layers, indicating different periods of occupation (Figures 6 and 7).The subsurface soil-formation (dark brown part) occurred from newly transported soil from a depth of 35-40 cm, which was subjected to a wide range of vegetation and slight to moderate weathering of the parent material.The absence of appreciable quantities of illuviated clay, organic matter and/or iron compounds, with medium-and fine-textured materials derived from a wide range of rocks, mostly in colluvial or aeolian deposits, contributed to its formation.The section showed a complex situation.

| Multivariate analysis
PCA (with its corresponding components designated as PC1, etc.) conducted on the elemental content data has unveiled intriguing patterns.In this section, we exclusively present commentary on the first three components, as they account for a significant proportion of the observed variability.Comprehensive results for all components, along with accompanying tables and figures, can be found in Figure 13 and Supporting Information S1: Figures S1 and S2.The first three components, PC1, PC2 and PC3, elucidated 47%, 22.23% and 11.81% of the total variability, respectively.PC1 showed significant associations with Al, K and Fe (positively), while demonstrating negative correlations with Mn and Zr.PC2 showed predominant correlations with Sr, As and Pb (negatively).Conversely, PC3 showed primary associations with Si (positively) and LE (negatively), as summarized in Table 2.
The elemental analysis reveals significant correlations, with positive associations observed for aluminium, potassium and iron, while adverse correlations are predominantly evident with zirconium and manganese.These empirical findings permit an inference regarding the inherent origins of the parent material in the study area.Specifically, aluminium, commonly associated with natural sources, tends to co-occur with clay minerals.Conversely, elements such as zirconium show stronger associations with mineral and sandy fractions.This distinction provides a preliminary demarcation between soil patterns stemming from geological underpinnings and those attributable to surface soil characteristics.Within PC2 (see Figure 13), noteworthy linkages emerge among strontium (Sr), arsenic (As) and LEs, which suggests potential anthropogenic influences.
Although these associations do not unequivocally signify a natural provenance, they plausibly indicate human activities spanning recent centuries, possibly involving atmospheric deposition.PC3 (see Supporting Information S1: Figures S1 and S2) further discerns silica from LEs, underscoring the dichotomy between organic matter and the mineral constituents within the soil.This analysis elucidates varying patterns and heterogeneity across the data sets.The bar plot of the PCA underscores abrupt transitions along PC1, indicative of deeper horizons predominantly influenced by geological factors, while the subsequent other components (PC2-PC4) align more closely with topsoil dynamics.PC1 effectively segregates profiles from topsoil and subsoil, emphasizing its capacity to distinguish between these soil strata.PC2 shows a negative association with Sr, As and LE, implying that more pronounced negative values correspond to increased concentrations of these elements.Predominantly, these negative values are prevalent in topsoil samples (see Figure 13), suggesting a potential legacy of atmospheric deposition.It is worth noting that Sr and LE may also have geological origins, but their more conspicuous presence in topsoil samples relates to contemporary pollution sources.This overlapping pattern in deeper soil layers can be attributed to various factors such as bioturbation, erosion, substantial leaching due to erosion, cryoturbation and the acidic nature of the soils.These conditions facilitate the translocation of these elements, depending on the specific soil profiles.Intermediate layers may contain sandy soil textures that facilitate element transport (see Figure 13), while lower horizons may accumulate higher levels of organic matter, leading to increased element retention, potentially indicative of recent pollution events.

| Grain size and magnetic properties
The grain size distribution within the study sections is variable.This is not so apparent in significant trends corresponding to the   Since the 11th century, substantial changes appeared across Europe, and also thus in the historical Czech lands.Development took place in economic, social and cultural spheres, and contemporary changes led to the human occupation of higher altitude areas and the foundation of many new villages.This process was usually run by a locator (i.e., development manager) who measured plans of the new settlement layouts and their agricultural hinterland.The peasants could only use the land that was measured for them.Regular visible boundaries of parcels of land started to arise, and these were constrained by the terrain of the landscape and the layout of villages (Klápště, 2005(Klápště, , 2012)).Environmental conditions, position or way of farming primarily drove the plan of villages and the layout of plužina.The most typical shape for villages in the foothills was a linear plan along the stream and their field systems having the shape of long strip fields, so-called croft plužina.The occurrence of croft plužina is typical for the submontane landscape of northern/northeastern Bohemia, as this type is often bound to hilly terrain (Fanta et al., 2022; Figure 1); however, it represents a common central European type of agricultural setting in the landscape.In the historical Czech lands, it is also distinctive for the area of the Jeseníky Mountains and the Zlatokorunská vrchovina Highlands (Šitnerová et al., 2020) or for the Šumava Mountains in the southern part of Bohemia (Beneš et al., 2022;Houfková et al., 2015).
There is a high probability that peasants were presented with this parcellation process of agricultural hinterland.A good knowledge of boundaries was indispensable: knowing which parcels they could take care of according to the system of medieval law (Razim, 2022).The boundaries delineating separate lands were made by earthen hedgerows, later covered by shrubs or trees, or some form of stony walls that could be overgrown by woody vegetation.These hedgerows were constructed during the foundation of the plužina field system.
First, the stones from the fields were collected into piles along the line of land boundaries.The stones were used as the basis for the construction of hedgerows.Such construction was discovered during the archaeological research into plužina field systems at Valštejn (Šitnerová et al., 2020).It seems also that the plužina field systems around Debrné were constructed using the same principle because and bioturbation higher degree of bioturbation; intensive leaching geodic nodules (Figure 7c); intensive leaching Notes Strange orientation of mica around pores or former pores-the main point of interest in this T-section is the unusual orientation of mica around pores or former pores, which may result from shrinking The thin section taken from 67 cm (Figures 6c and 7b,d,f) revealed the formation of lixisols from the weathered geological substrate Interestingly, a higher degree of bioturbation; intensive leaching; and a greater presence of Fe/Mn nodules were observed at 45 cm compared to the 67 cm thin section.Possible human or natural colluvium process The platy microstructure in the thin section at 25 cm may have been caused by heavy mechanization or freezing-thawing cycles: particularly cryoturbation due to the presence of thin rims of Fe-rich matrixes in voids.It might be the result of the Little Ice Age High presence of organic matter.The orientation of the matrix; it is for sure due to the press; similar to that in hand-made ceramic.It is obvious from the context that there should be a mix and matrix with cow excrement, but no faecal spherulites were observed.
Due to the unfavourable conditions for their preservation (Figures 6 and 7 | 443 remnants of the relevant stone accumulations were found during the excavations.Stones, stony and earthen walls with woody vegetation served as a boundary of the land, but also had ecological significance (Sklenička et al., 2009), as they retained water and defended against soil erosion.The trees growing in the hedgerows were also used as a source of wood, and they co-created specific ecological habitats.The most common species of hedgerows were hawthorn (Crataegus monogyna), hazelnut (Corylus avellana) and blackthorn (Prunus spinosa) (Baudry et al., 2000;Mccollin et al., 2000).In the case of the Debrné, the presence of shrubs of the genus Rubus and Sambucus has been demonstrated, which is in accord with the archaeobotanical records of these botanical families at Valštejn (Šitnerová et al., 2020).76,952.17 220,207.25 22,649.46 182.54 50.43 1011.15 33,406.82 154.17 51.86 238.80 46.28Both radiocarbon and OSL dating methods were used in testing their suitability for establishing a chronology of soil deposition in the studied trenches.The results confirmed the expectation that it is optimal to use charred macroremains of short-living plants (Supporting Information S1: Table S2), obtained from the systematic flotation of sediment from the mechanical layers in the parcel boundaries.The  Trench S1 had more or less the same formation process but with younger material compared to trench S2.
The formation of the earthen field boundaries in the medieval croft plužina field systems was a result of deliberate human actions and land management practices.These boundaries played a crucial role in organizing and maintaining agricultural fields during the medieval period.
| 445 The question that remains to be answered is whether the accumulation layers at the edge of the earth terrace-like terrain steps are the result of long-term ploughing, which could have gradually deposited soil over time, or whether some of these layers were the result of the movement of soil material at the time of construction of the plužina system, potentially indicating deliberate earthwork activities during its establishment.Either way, it is clear from the profiles of the terrace-like formations that a significant amount of material at the base of the terrace was excavated by humans.This excavation serves as compelling evidence for anthropogenic intervention, suggesting that it was extracted for the specific purpose of land levelling, a crucial aspect of landscape modification and construction planning.In summary, earthen field boundaries in plužina were created through a combination of land preparation, agricultural practices and cultural traditions.They played a vital role in organizing and maintaining agricultural land during this historical period.

| XRF elemental content and multivariate analyses
The PCA results (Figure 13 of the plužina.Nevertheless, all classifications identify these fundamental methods of the spatial POLEDNÍK MOHAMMADI ET AL. | 429 organization of parcels: (a) segmental or blocky plužina (square-like irregular plots), (b) sectional plužina (narrow strips of land, organized parallel to each other, arranged into larger blocks) and (c) croft or longitudinal plužina (very long parallel strips belonging to each farmstead).Each type reflects specific environmental conditions and/

F
I G U R E 2 (a) Location of former Debrné village in Central Europe.(b) Prehistoric and medieval/early modern archaeological sites are recorded in the Archaeological map of the Czech Republic.Source: The Institute of Archaeology, Czech Academy of Sciences, Prague.(c) Position of trenches in the landscape geomorphology (red dots); the altitude varies between 450 and 679 masl.(d) Present-day landscape of the former Debrné village with the visible hedgerows of the plužina.Photo by T. Jůnek; layout and picture processing by J. Bumerl.

F
I G U R E 6 Micromorphological documentation of the main differences between soils accumulated in different timespans at Debrné, trench S2: (a) Moderate developed Fe rims in sample from 25 cm.(b) Example of the phytolith located in phosphatic and organic matter-rich granular microstructure of sample at 25 cm trench S2 with the small size of microcharcoals (PPL).(c) Appearance of limonite neoformation (red arrow) and Fe/Mn concentrations (blue arrow) with the presence of plant residues (yellow arrow) at the depth of 67 cm (PPL).(d) Cross-section of stems (yellow arrow).The brown border corresponds to the epidermis.The groundmass is made of a thick and convoluted clay coating and Fe/Mn infillings formed by multiple laminations related to different translocation phases.Some of the contacts between laminations correspond to erosive surfaces and limonite neoformation (red arrow) at the depth of 110 cm (PPL).(e) Fragments of wood charcoals (yellow arrow), opaque in PPL with very fine and coarse porosity patterns, respectively.The microstructure is fine granular and related to biological activity with ironbearing nodules with undifferentiated internal fabric and sharp boundaries (red arrow) at the depth of 110 cm (PPL).(f) Alteromorphic nodules: This type of nodule is usually the product of weathering (red arrow).These nodules are also characterized by pseudomorphosis of mineral or organic materials with the presence of microcharcoal (yellow arrow) at the depth of 45 cm with the bioturbated matrix (PPL).PPL, planepolarized light.POLEDNÍK MOHAMMADI ET AL.
The radiocarbon date from 70 to 80 cm in trench S2 produced an age estimate of 1025-1175 A.D., which precedes the first written source about Debrné.The radiocarbon dating from the upper level of the S2 falls in the range 1492-1653 A.D. An OSL sample was taken at 70 cm in trench S2.The result is illustrated in the graphs in Figures 9 and 12 and shows that the sediment was last exposed to sunlight during the third century A.D., that is, in the Roman Iron Age period.
Trench S2 showed a macroscopic distinction of up to six discernible layers characterized by their field colour and material composition.These layers spanned a range of 2.5YR hues, including 3/3 and 3/4, and featured textures that varied from clay to sandyclay and silty clay.Archaeologically, these layers have been linked to different occupational and geomorphological events during the 12th century (occupational event as determined by radiocarbon dating) and the third century A.D. (the sedimentary records, as determined through OSL dating at a certain depth of 70 cm, were not affected by any human activities; refer to Figures4 and 8).The soil horizons in trench S2 showed distinct morphological characteristics, including easily demarcated horizons with wavy or smooth topography.The 210 Pb and 137 Cs values indicate that the sediment in the profile does not show signs of disturbance.The 210 Pb values are near the detectable limit.The uppermost layer at trench S2 (the Ah) (Figure 8, b1), up to a depth of 50-55 cm, was enriched with organic matter and shows a dark reddish-brown material colour (2.5YR 3/3).Just below the Ah was the topsoil (AhB) (Figure 8, b2) extending from 55-60 to 100-105 cm, which also shows a dark reddish-brown material colour (2.5YR 3/3).This layer contains microcharcoal and some alluvial deposits.Moving further down, we encounter the weathered B-horizon.It features a dark reddish-brown material colour (2.5YR F I G U R E 12 Differences between archaeological dating and dating by written sources ('historical dating').The black curve with grey confidence intervals represents a predictive model describing the relationship between the archaeological dating and the historical one (the blue and green thin segment lines represent the data of 527 settlements used for the derivation of the predictive model).The thick red segment line represents the interval of Debrné's oldest AMS 14 C calibrated date from a charred annual plant seed in the S2 trench-a lower medieval layer.Source: Adapted from Fanta et al. (2020).
3/3) and contains a moderate amount of clay, characteristic of cambisols.The next horizon is marked by the illuvial accumulation of clay (Bt) (Figure 8, b3) from approximately 105 to 130 cm with sandyclay, clay, silty loam and silty clay texture.It presents a dark reddishbrown material colour (2.5YR 3/4) and includes transported microcharcoal from upper layers and small stones.This layer indicates the formation of lixisols and possibly the removal of the Ah.Below the Bt horizon, we the C1 horizon, which is dark reddish-brown (2.5YR long-term pedological transformations, human influence or change in provenance in trench S1, but does show up more or less in trench S2 around the 70-80 cm depth.The majority of the measured samples show an upward trend in the coarse fraction content, except for those collected from the upper 70 cm section, where only sandy fractions were identified in these particular samples.The silty fraction prevailed in only a few samples with a sandy fraction.This can possibly be the result of cryoturbation or partial bioturbation, either of which can reflect a different material provenance or post-depositional changes.The magnetic susceptibility values, serving as an indicator of grain size distribution (see Supporting Information S1: Figure S3), demonstrate significant variability across all examined sections.Notably, in this case, there is a clear lack of dependence on grain size distribution.The most enhanced susceptibility was documented POLEDNÍK MOHAMMADI ET AL. | 441 T A B L E 1 Micromorphological description of soil samples at Debrné, trench S2.
at the depths of 70-80 and 40-60 cm, which can reflect the transported soil material or colluvium.The uppermost layers of trench S2 suggest the presence of a medieval plowed Ah at a depth of approximately 30 cm and a plowed colluvium from the slopes above contained by the air pollution of the last centuries.The deeper part contained a weathered geological substrate in which, in individual sections, the lowermost values of magnetic susceptibility were usually detected at the very bottom of the sections in the cryoturbated geological substrate.5 | DISCUSSION 5.1 | Foundation, dating and development of the agricultural field system of Debrné

F
I G U R E 13 Bar plot of principal component 1 (PC1) versus principal component 2 (PC2), trench S1.T A B L E 2 Results of principal component analysis for 46 samples.
Figure12).Approximately the same time lag was caught during the research of the abandoned village Malonín in the Šumava Mountains at the beginning of the high medieval period(Houfková et al., 2015), and already no time lag in the case of the early modern field system of Valštejn in the Zlatokorunská vrchovina Highlands in the Czech part of Silesia at the beginning of the 17th century(Šitnerová et al., 2020).The foundation of the parcel boundaries in Debrné is ca.900 years old and, even with its subsequent extension, has survived to the present day virtually unchanged.The parcels of arable land with their fixed earthen boundaries, often with a terrace-like character in the sloping terrain, were deliberately built during the establishment of the village by way of large-scale construction projects.The older medieval dating of the parcel boundaries has been confirmed not only at Debrné but also at other sites that have been investigated using methods of environmental archaeology(Beneš et al., 2022).This shows conclusively that most of the earthworks were carried out at the time of the foundation of the medieval settlement.A common feature of the older dating of the foundation of fields and parcel boundaries in Malonín and Debrné than that given by written historical sources is that the charred remains of short-lived plants were used for 14 C AMS.The use of annual plants for radiocarbon dating has been recently discussed in the case of the Holocene history of common human and vegetation development in Czechia(Kolář et al., 2022).The use of short-lived plants obtained by the archaeobotanical method of systematic layer flotation seems to be essential in our research on medieval field systems.Both methods (AMS, OSL) are common for dating parcel boundary structures, terraces and stone walls.It seems very probable in this case that the OSL shows that the third century A.D. meant a period of sedimentation of the slope deposits, and not the period of creation of the hedgerows.The basis for our conclusion lies in the lack of any settlement activity during the Roman Iron Age period within this particular part of the Trutnov region.This area, characterized by minimal habitation until the Middle Ages, underwent a process of colonization during the 13th century.This colonization was spearheaded by the Švábenští family, a Moravian aristocratic lineage, under the rule of the Přemyslid royal dynasty.

5. 2 |
How were the earthen field boundaries formed?The buried soil within the Debrné site was preserved irregularly.Parent material that is partly eroded is revealed under the topsoil of the earthen wall and represented by sands, sandy loams and clay, in some cases underlain by moraine loams.All this makes it difficult to reconstruct the component composition of the soil cover when reconstructing the original landscape conditions that can generally yield good agricultural land that can be used intensively.The soils at Debrné (trench S1 and S2) were classified as lixisols and cambisols.These soils have the following set of genetic horizons, which are typical cambisols with at least an incipient subsurface soil-formation.Transformation of the parent material is evident from the structure formation and mostly brownish discoloration, increasing clay percentage and/or carbonate removal (IUSS Working Group WRB, 2014).Based on the field excavation in Debrné (trench S2), it was obvious that the area is affected by heavy leaching and illuviation, which can be due to erosion and transfer of eroded material.These situations are most obvious from the micromorphological samples from the depths of 47, 67 and 110 cm, in which the presence of humanaffected features, such as microcharcoal, could be because of material having been transported by leaching to a deeper part.The presence of Fe/Mn nodules and oxidation/reduction can be perfect proof for these claims.Another fact could be the presence of biological activity such as bioturbation, which affects the microstructure of the soil in many parts of the soil profile.Based on the micromorphological thin sections, field observations and data set, it is possible that the area is subjected to heavy mechanization or freezing-thawing cycles, and especially cryoturbation, due to the presence of the thin rims of an Fe-rich matrix in voids.It seems that in trench S2, we have two phases of soil-formation.In the first phase (deeper part, 50-130 cm), due to heavy mechanical activity or erosion, the Ah of this part was removed, and then in the next phase, the horizon started to form from some transported material under a wide range of vegetation with some slight or moderate weathering of the parent material and by the absence of appreciable quantities of illuviated clay, organic matter and aluminium and/or iron compounds.Medium-and fine-textured materials were derived from a wide range of rocks, mostly in colluvial, alluvial or aeolian deposits at a depth of 0-20 cm.At a depth of approximately 25-40 cm, with a moderate amount of clay, the soil order in the deeper part could be lixisols.The reddish colour results from rubification brought about by dehydration of iron compounds in the long dry season.The agric surface horizon lacks clear evidence of clay illuviation other than a sharp increase in clay content over a short vertical distance with low levels of available nutrients and low nutrient reserves (probably the removal of the Ah and the E-horizon).The parent material was unconsolidated, strongly weathered and had strongly leached, finely textured materials.
and Supporting Information) indicate that phosphorous levels were lowest in the samples.Elements like Al, Fe, Mn and LEs showed strong correlations, possibly linked to such factors as region, soil type and human activity.In PC1, positive associations of aluminium (linked to clay minerals in natural sources), potassium and iron and negative associations of zirconium (linked to mineral and sandy fractions) and manganese can be linked to the natural origin of the parent material in the area.Aluminium tends to be associated with clay minerals in natural sources, whereas elements such as zirconium are more commonly linked to mineral and sandy fractions.In summary, distinguishing between geological background and topsoil-related soil patterns can be challenging.PC2 shows negative connections with Sr, As and LE, indicating higher amounts of these elements in samples with more negative values.These negative values are predominant in the topsoil, possibly suggesting human activity influence, such as atmospheric deposition over the last few centuries.Notably, while PC1 and PC2 in profiles S1 and S2 show rough similarities (Figure13), PC3 with a positive correlation with silica and a negative correlation with light elements reveals greater element diversity that can influence topsoil and subsoil (Supporting Information S1: FiguresS1 and S2).Profile S2 demonstrates unique components and directions for each sample, indicating the strong influence of silica content in the topsoil for profile S1 and more factors affecting light elements in profile S2 (Supporting Information S1: FiguresS1 and S2).Consequently, profile S2 is broadly associated with organic matter and light elements, highlighting significant differences between the two trench profiles.We observe both patterns and variations in all data (trench S1 and S2).However, a challenge arises as similar patterns emerge in deeper layers.This similarity may be attributed to factors such as bioturbation, extensive leaching from erosion, cryoturbation and soil acidity, all contributing to the movement of elements within soil profiles.Some geochemical signals may migrate through the soil profile, with a sandy layer in the middle potentially facilitating this element's transport.Principal component 4 (PC4) highlights a noticeable contrast: the current topsoil is significantly more compact, whereas lower layers are strongly influenced by geological factors.This suggests that trench S1's profile resembles that of trench S2, albeit without a bottom layer.Trench S1's profile records 16th-century rebuilding activity, coinciding with the expansion of the field system.In trench S2, radiocarbon dating of charred seed macroremains indicates an origin in the 12th century A.D., aligning with both the profile's history and OSL dating from the third century A.D. The difference between the dates is probably because the activity of the medieval founders of the village affected the slope layer deposited there almost 1000 years earlier.Radiocarbon dating corroborates the age of the earthen field boundary in trench S2, confirming its connection to the initial construction of the medieval system.6 | CONCLUSIONGeoarchaeological research in the Debrné area has revealed a complex formation and chronology of historical earthen field boundaries by combining several methods.The sedimentary profile within Debrné, trench S2, showed notable variations in both macroscopic and archaeological aspects, offering valuable insights into the site's history and soil characteristics.On a macroscopic scale, the accumulation of soil material was stratified into as many as six distinguishable layers, which were characterized by distinct and abrupt boundaries.In terms of archaeology, these stratigraphic layers were associated with discrete occupational events that occurred during the 12th century.Furthermore, the soil horizons within trench S2 showed well-defined horizons with either undulating or smooth topography.The uppermost layers showed hues of dark brown, and dark yellowish brown, accompanied by textures ranging from sandy-clay to clay and silty loam.Microscopic analysis of the samples unveiled distinctions among the archaeological layers, indicative of various phases of human occupation.The soil profile also featured notable pedo features, including Fe/Mn nodules, organic fractions and passage features.Moreover, thin sections of the soil revealed the existence of distinct soil orders, notably cambisols, and lixisols, each characterized by unique microstructures and grain sizes.Detailed observations of mica orientation, bioturbation and other pedo features contributed valuable insights into the intricate soil-formation processes and underlying geological substrates present at the Debrné site.This comprehensive examination of both macroscopic and microscopic features in the sedimentary record at Debrné, trench S2, not only enhances our understanding of the site's history but also provides critical information regarding the complex interplay of soil-formation processes and geological conditions in this archaeological context.The archaeological examination unveiled intriguing findings, including stone drainage systems, buried medieval soils and various formation elements.In trench S1, which showed a relatively more recent construction date, a significant number of larger stones were unearthed.Radiocarbon dating of trench S1 pointed to the terrace's origins in the latter half of the 16th century.In contrast, trench S2, ); reference sample Abbreviations: Ab, angular blocky; Am, amphibole; Bl, blocky; Bt, biotite; Cal, calcite; Cdp, compound packing; Chm, chamber; Chn, channel; Cl, chlorite; Cr, crystallitic; Fsp, feldspar; Gr, granular; Ma, massive;Mc, microcline; Mi, mica; Mo, monic; Op, opaque; Ot, other; Pl, plagioclase; Po, porphyritic; Pyr, pyroxene; Qu, quartz; Rf, rock fragment; Sb, subangular blocky; Ser, sersericite; Sp, simple packing; Srp, serpentine.POLEDNÍK MOHAMMADI ET AL.