Analysis of prehistoric brown earth paleosols under the podzol soils of Exmoor, UK

The deforestation of the upland landscapes in southwest Britain during prehistory is an established archaeological narrative, documenting human impacts on the environment and questioning the relationship of prehistoric societies to the upland landscapes they inhabited. Allied to the paleoenvironmental analyses of pollen sequences, which have provided the evidence of this change, there has been some investigation of prehistoric paleosols fossilized under principally Bronze Age archaeological monuments. These analyses identified brown earth soils that were originally associated with temperate deciduous woodland, on occasion showing evidence of human impacts such as tilling. However, the number of analyses of these paleosols has been limited. This study presents the first analysis of a series of pre‐podzol brown earth paleosols on Exmoor, UK, two of which are associated with colluvial soil erosion sediments before the formation of peat. This study indicates these paleosols are spatially extensive and have considerable potential to inform a more nuanced understanding of prehistoric human impacts on the upland environments of the early‐mid Holocene and assess human agency in driving ecosystem change.

In comparison, Exmoor contains relatively few stone monuments, which are composed of nongranitic, smaller (<0.5 m) stones, known as "miniliths" (Gillings, Pollard, & Taylor, 2010). Likewise, the archaeological record of both Dartmoor and Bodmin Moor contains numerous Early Bronze Age barrows and cairns, Middle Bronze Age land divisions (reaves), Middle Bronze Age enclosures, some containing multiple roundhouses, and later prehistoric enclosures, such as hillforts (Johnson et al., 2008;Newman, 2016). Early Bronze Age barrows and cairns are also found on Exmoor, such as the Chapman Barrows and cairn groups, for example, Robin and Joaney How (Riley & Wilson-North, 2001, pp. 32-40). In contrast, Exmoor has a relative lack of surviving prehistoric land divisions, although some are suspected at Codsend Moor (Francis & Slater, 1992), Hoar Moor (Francis & Slater, 1990), and Chetsford Water (Riley & Wilson-North, 2001, p. 54). Evidence for prehistoric settlement is present on Exmoor, first in the form of house platforms and several hut circles which, although currently undated, may be analogous to those shown to be of Middle Bronze Age date on Dartmoor and Bodmin Moor. Settlement also occurred in the form of small enclosures (<1 ha), some of which are likely to be Iron Age, although others as at Holworthy (Green, 2009) may have Bronze Age origins.
Iron Age activity on Exmoor is also demonstrated by seven hillforts located on the edge of the moor overlooking river valleys (Riley & Wilson-North, 2001, pp. 56-64).

The disparity in the visibility of the archaeological records between
Exmoor and the other upland areas of the southwest peninsula has led to a difference in their investigation, which also extends to the study of their paleoenvironmental and geoarchaeological contexts. On Dartmoor, the field systems associated with prehistoric enclosures and houses were extensively mapped (Fleming, 2008), with some allied geoarchaeological and paleoenvironmental investigations (Balaam, Smith, & Wainwright, 1982). Detailed paleoenvironmental analysis has been undertaken on Dartmoor demonstrating some use of fire in the Mesolithic period for vegetation clearance and a late Mesolithic oak, elm, and hazel dominated woodland. It was previously interpreted that significant woodland clearance occurred during the late Neolithic-Early Bronze Age, with a largely cleared landscape established by the time of reave construction during the Middle Bronze Age between 1700 and 1300 BC (Caseldine, 1999;Caseldine & Hatton, 1996;Wilkinson & (UB821, calibrated OxCal v. 4.3), with human activity visible in the pollen record interpreted from the Neolithic onwards, although increased human activity is noted in the pollen from between 1044 cal BC (UB-819 calibrated OxCal ver. 4.3) and 426 cal AD (UG-816, calibrated OxCal ver. 4.3;Merryfield & Moore, 1974). Straker and Crabtree (1995) also analyzed peat deposits at the Chains and suggested a date of c. 3000 BC for peat inception, but attribute this to a climatic driver, interpreting no human impact within the pollen spectra before 1000 BC. On Codsend Moor, Francis and Slater (1992) provide a date for peat inception of no earlier than 470 BC and suggest a model of human-induced forest clearance for livestock grazing during the Mid-Late Bronze Age, with Late Roman pasture and arable seemingly well represented. More recent research on Exmoor has focused on pollen analysis from spring mires that has provided localized sequences next to the monumental complexes of the Seta and Five barrow cemeteries (Fyfe, 2012), demonstrating a Middle Bronze Age date for the major paleoecological transformation of this landscape. Pollen analysis at Molland on the southern edge of Exmoor (Fyfe, Brown, & Rippon, 2003) records a short-lived Early Neolithic woodland disturbance, some woodland clearance from the Early Bronze Age, but intensive woodland clearance from the early Iron Age, creating a cleared pastoral landscape. In the nearby Blackdown Hills, pollen analysis shows some Late Bronze woodland clearance and Iron Age cereal cultivation, with little evidence for Neolithic woodland clearance (Brown et al., 2014).
This removal of woodland in prehistory is intimately linked to the degradation of the original early Holocene brown earth soils within these upland systems. However, in contrast to the paleoenvironmental analysis of pollen data for Exmoor and its surrounding areas, the geoarchaeological study of soils and sediments has been limited.
The transition from the original loess derived upland acidic brown earth woodland supporting soils of the early Holocene into podzolic soils is the product of acidification and waterlogging, although as Moore (1993) discusses a variety of factors are likely to contribute to this, including local topography, contemporary land use, geology, and climatic variability. Acidification and waterlogging were exacerbated by tree cover removal, with grasses and herbs like heather reducing the rate of evapotranspiration, and plant litter accumulation at the soil surface increasing localized anaerobism. This facilitated clay breakdown and mobilization of sesquioxides, causing brown earth soils to degrade into stagnogley podzols and peats on the higher moors (Dimbleby, 1962;Duchaufour, 1982, pp. 112-121;Macphail & Goldberg, 2018, pp. 122-134;Maltby, 1995). Across the southwest peninsula, the chronology of brown earth soil deterioration and transition into podzol systems is poorly understood, especially in relation to the dynamics and activities of past human societies. The analysis of sediments beneath Saddlesborough Reave, Dartmoor, demonstrated the soils at this locality had already undergone severe degradation, with a peaty topsoil, Eag horizon, and iron pan over a podzolic B horizon present before reave construction in the Middle Bronze Age (Balaam et al., 1982;Macphail, 1980). At Chysauster, Cornwall, an early Holocene brown earth soil that had developed on late Devensian loess above granite was recognized. This paleosol under a cairn had been affected by some acidification before burial, exacerbated through tilling, demonstrating soil degradation before podzolization was occurring in the Early Bronze Age with significant colluviation during the Bronze Age/Iron Age (Macphail, 1987;Smith, 1996). Maltby and Caseldine (1982)  However, preburial this soil had undergone podzolization, with leached iron and clay microfabrics in the upper Ah/Ea horizons, which contained significant amounts of charcoal. The podzolization of this soil was related to early Neolithic woodland clearance, causing soil instability and initiating podzolization, although the process of podzolization was halted due to burial under the bank shortly afterwards (<300 years; Macphail, 1989). Brisbane and Clews (1979)  Canal, whereby a mineral soil that predated the peat formation was found underneath the bank of a C19th canal (Crabtree & Maltby, 1975). However, given the extent of the southwestern upland areas in the UK, combined with the rich prehistoric archaeological records they preserve, these studies represent a small sample of analyses.
The synthesis of these studies also demonstrates a clear potential to directly elucidate the chronology and impacts of human societies on past landscapes within this region. This paper then presents the geoarchaeological identification and analysis of a series of paleosols across Exmoor resulting from site investigations during the Exmoor Mires Project.

| Background to this study
The Exmoor Mires Project was part of the Upstream Thinking initiative undertaken by Southwest Water from 2011 (South West Water, 2020), which focused on the restoration of the hydrological function of mires as a flood prevention strategy and improving water quality across the catchment (Bray, 2015). This rewetting involved the blocking of historic drainage ditches, requiring heavy plant to traverse the moor along access tracks and necessitating site-specific archaeological mitigation strategies within impacted areas. These interventions used geophysical surveys and targeted excavation to understand the archaeological remains, coupled with geoarchaeological investigations of soils and sediments, to understand the pre-peat deposit sequences. The podzol soils on Exmoor have been shown to contain relatively thin peat deposits (<c. 1 m; see below) away from the spring lines and blanket bogs, and although excavations have revealed archaeology such as Mesolithic flint scatters under the peat (e.g., Hawkcombe Head; Gardiner, 2007), and standing stones now surrounded by peat but cutting an earlier deposit sequence (Gillings et al., 2010), the systematic investigation of the pre-peat podzol landscape of Exmoor has been lacking. This paper details the results from the multiproxy geoarchaeological investigation of three sites across Exmoor, investigating the pre-peat sediment sequences associated with archaeological remains at Wintershead

| Field sampling
Excavations were undertaken using trenches to investigate archaeological features. The geoarchaeological sampling of sediment sequences collected tins from trench sections for laboratory analysis. Samples were collected using a square plastic drainpipe, with one edge cutoff. The drainpipe was placed over the section and labeled, photographed and recorded on the section drawing, before sample removal. The sample was wrapped in clingfilm and black plastic, before being placed in cold storage.

| Laboratory subsampling
Samples were cleaned, photographed, and logged. Each sample was continuously subsampled on 1 cm interval removing c. 10 g of sediment. The 1 cm subsamples allowed sediment variation both within and between contexts to be analyzed, integrating the analysis with the field excavation data. Spot samples for pollen analysis were collected from each context, with their location recorded. The remaining undisturbed sediment was retained for soil micromorphology. The 1 cm subsamples were oven-dried at 40°C. When dry, each subsample was homogenized in a ceramic pestle and mortar, and fractionated using a 2 mm sieve. The <2 mm fraction was weighed and discarded, with the ≤2 mm fraction retained for analysis.

| Analysis of fine sediment fraction
The fine sediment fraction was analyzed to determine sediment composition using a Malvern Mastersizer 2000 laser analyzer, using a Mie scattering model (Malvern, 2005). Each subsample was disaggregated through adding 5 ml of sodium hexametaphosphate (Calgon) to a heaped spatula of sediment (c. 1 g), which was agitated on a platform rotary shaker at 175 rpm for a minimum of 1 hr. Each subsample was analyzed using Basic Ultrasonic Method, making three measurements, with a mean value calculated. All data were exported from the Malvern Mastersizer using the Wentworth scale, a Phi classification of sediment sizes range (Table 1), and this nomenclature is used throughout.

| Organic content
Loss on ignition was used to measure the organic content, which is a useful proxy for the identification of paleosols (Canti, 2015). Ceramic crucibles were oven-dried at 100°C for 24 hr before weighing.

| Magnetic susceptibility
Magnetic susceptibility was used to identify evidence of heating, as well as topsoil inwashing, with both processes enhancing magnetic susceptibility values (Goldberg & Macphail, 2006, pp. 350-352). The magnetic susceptibility of each subsample was measured using a Bartington MS2B magnetic susceptibility meter with the reading calibrated to the mass of the sample, using 10 ml pots. The sample sequence required a blank zero measurement before the sample was added to the meter and the magnetic susceptibility measured for 5 s, before removal and a further blank zero measurement, to calibrate for drift. The sample measurement was mass-specific.

| Soil micromorphology
Thin sections were taken from within significant contexts, to identify brown earth fabrics, inclusions, and pedogenic features in the sediment sequences underlying the current podzol soil formations (Stoops, Marcelino, & Mees, 2018). Each thin section sample was impregnated with a clear polyester resin-acetone mixture; samples were then topped up with resin, ahead of curing and slabbing for 75 × 50 mm-size thin sections. Thin sections were further polished with 1,000 grit papers and analyzed using a petrological microscope under plane polarized light, crossed polarized light, oblique incident light, and using fluorescent microscopy (blue light), at magnifications ranging from ×1 to ×200/400.

| Pollen
Standard techniques for concentration of the subfossil pollen and spores were used on subsamples of 1.5 ml volume (Moore & Webb, 1978;Moore, Webb, & Collinson, 1991

| Presentation of data
With the analyses completed, the data were entered into an Excel spreadsheet, before exporting to SPSS for the drawing of line graphs.
Graphs drawn in SPSS were exported in Adobe Illustrator, and added to the sample logging sheet, with the context boundaries drawn over the graphs. Each context was then described and the data from the soil micromorphology and pollen analyses were integrated.

| RESULTS
The results from each site will be presented individually before a

| Wintershead
Five trenches were excavated at Wintershead EWH13 ( Figure 2), positioned to investigate anomalies identified by the gradiometer survey.
Within Trench 4, three intercutting pits were found cutting context (420) Context (402) lower interpretation: Given the spikey distribution of the particle size data and the presence of frequent small stones, (402) lower is interpreted as colluvial sediment. The identification of burnt quartzite in the thin section, complements the increase in magnetic susceptibility and indicates burning, correlating with colluvial soil disturbance and deposition. As this subunit is colluvially derived, the pollen data should be treated with caution, but as with the previous contexts indicates a cleared landscape, with some patches of woodland and acidified environments within the pollen catchment.  (420) and (407) containing welded soil microfabric/total excremental microfabric). (b) Photomicrograph of the mixed boundary of (402) lower and silty peat (402) upper. Note, probable charred root (center) probably relates to management by fire (or lightning strike) of the newly developed wet ground soils (humic silt could be colluvial). PPL, frame width is~4.62 mm. (c) Photomicrograph of (420) with compact soil ("welded" soil microfabric/total excremental microfabric) due to an earlier history of brown earth soil earthworm working. PPL, frame width is~4.62 mm.   Plantago coronopus (plantain) type  (407) and (420), supporting the interpretation of deposition of tephra shards postdeposition of the colluvium.

| Wintershead summary
The analysis of the sediment sequence from Wintershead provides a snapshot of the evolution of the landscape at this locale. Contexts (420) and (407)   Mesolithic including a thumbnail scraper, a microcore, a microburin, a denticulate and a narrow retouched blade (Gardiner, 2019). In addition to the sediment analyses, three pollen subsamples were analyzed from <FW2> and these had a low level of pollen diversi ty and high polypodium numbers, indicating poor preservation and differential survival (Figures 6 and 7; Tables 6-9).
Context (2-05b) is a light brown-orange silt, with sand and clay. Clay This deposit is somewhat anomalous in terms of ferric stagnogley podzol formation. The sediment is partly derived from a low energy colluvium, containing smaller particle sizes (clays and silts). Evidence of burning was noted, possibly a consequence of the overlying podzol mire management, explaining the magnetic susceptibility spike and burnt quartzite

| Context (2-02a)
Context ( Context (2-02a) interpretation: An almost stone-free minerogenic peaty soil has formed (Oh horizon), a product of podzolization, with organic matter accumulation and contemporary inputs of silt and very fine sand. A moderate amount of bioworking has also taken place.
Magnetic susceptibility values remain low, although spikey, with no discernible trend. No soil micromorphology or pollen analysis was undertaken on this context.

Context (2-01) interpretation: This is the modern lower A horizon
of the podzol soil formed post the degradation of the brown earth.
The A horizon (minerogenic Oh) is relatively thin and organic rich.
While a peat, there is still some clay and silt fractions evident in the soil matrix.

| Lanacombe
At Lanacombe (ELN14) two trenches investigated a series of pits forming an ovoid enclosure (c. 80 m diameter long axis) and linear features, identified by the gradiometer survey ( Figure 8; gradiometer data not shown). Trench 4 excavated one of these pit features, with <LC05> sampling the sediment sequence, collecting 40 cm of sediment. Contexts (453) and (452)

Context
Depth on sample (cm) Soil micromorphology description Soil micromorphology interpretation (2-05a) 26-21 Heterogeneous and broadly horizoned with very dominant pale brown and dusty fine sandy silt loam, becoming dominant upwards, brown to dark dusty brown fine sandy silt loam, with few fine burrow fills with dark reddish-brown humic soil. It is poorly sorted with a silt, coarse silt, very fine sand matrix containing sand to coarse sand-size ferruginous soil clasts and common small stone-size (<12 mm) siltstone and quartzite fragments with 0.5-1 mm thick bleached rims. Rare birefringent arbuscular mycorrhizae fungal bodies, occasional fine roots (<0.75 mm) and rare trace fine charcoal. There are rare iron-rich fine clay void coatings and infills within relict iron-rich soil, occasional matrix intercalations and associated dusty clay void coatings and infills (argillic fabric), very abundant depletion features, gray matrix soil, bleached stone rims, likely occasional organo-sesquioxidic staining along burrows (hypocoatings and polymorphic soil), and rare patches of iron staining of "argillic fabric" soil, occasional becoming many (upwards) thin burrows, and rare becoming many, upwards, and very thin organomineral excrements Soil palimpsest with stony subsoil formed at the top of the Pleistocene Head, and displaying a trace of early Holocene argillic brown earth (Bt horizon) formation (2-02b) 21-14 Rare to trace becoming occasional fine charcoal (max 0.5 mm) above 20 cm. The profile continues to be heterogeneous upwards, with strongly mixed fine patches of brown to dark dusty brown fine sandy silt loam and dark reddish-brown humic soil, which becomes more dominant upwards. Common gravel and small stones (<18 mm), often with bleached rims and/or near totally bleached character are present, alongside many fine roots remains, rare birefringent arbuscular mycorrhizae fungal bodies, occasional fine charcoal mainly in brown fine sandy silt loam soil patches/clasts. There are occasional matrix intercalations in brown fine sandy silt loam, very abundant relict bleached stones and brown silt loam that is iron-depleted, a rare trace of iron root staining, many weak organo-sesquioxidic/ organic staining of darker brown fine sandy silt loam, abundant thin burrows and burrow-mixing of brown and dark brown loam soil, and many very thin (and sometimes organic pellety) and many thin organomineral excrements A moderately iron-depleted colluvium formed of presumed Eb (A2) horizon soil, containing a small concentration of fine and very fine charcoal, and matrix textural pedofeatures suggesting muddy colluvial (possibly trampled?) formation. Later leaching, iron, and organosesquioxidic soil formation (weak Bs horizon) occurred, the last being introduced by burrowing from above. Upwards, the more humic but still stony fine sandy silt loam (Bs horizon) becomes dominant, although many small relict patches of fragmented fine sandy silt loam colluvium are present. The latter include fine charcoal (2-02a) 14-10.5 Very few gravel (<6 mm). The horizon is composed of very abundant amorphous organic matter and many fine roots, with occasional fine charcoal (<0.3 mm) and rare fungal materials are also present. A rare trace of ironstained root residues, occasional bleached rims gravel, abundant thin burrows, and abundant very thin and thin organic excrements (containing fine mineral material) An almost stone-free minerogenic peaty topsoil has formed (Oh horizon), through organic matter accumulation and contemporary inputs of silt and very fine sand. A moderate amount of bioworking has also taken place Context (452) interpretation: This is mainly Bs podzol horizon with mixing with humic pellety Bh horizon soil which becomes more dominant upwards, before being cemented into the iron pan. There is limited evidence of original brown earth soil profile, presumably A/upper B horizons that have been affected by podzolization.
Context (451) interpretation: This is a stony, peaty, topsoil mixed with leached E horizon, and weathered rock. The presence of podzolic B horizon soil was not noted. The mixed topsoil material contains only a small trace of charcoal.

| DISCUSSION
The analysis of these samples demonstrates the residual presence of brown earth paleosols under the current peat deposits of Exmoor.
The identification of paleosols with relict argillic brown earth fabric is consistent with previous pollen data that demonstrates a widespread temperate deciduous tree cover in the early Holocene that was subsequently cleared (e.g., Maltby, 1995;Merryfield & Moore, 1974).
The identified brown earth paleosols are clay to fine silt dominated, with the particle size data demonstrating a reduction in the fine sediment fractions in the upper parts of the sediment sequences, as clay and very fine silt become depleted due to a combination of clay translocation, acidification, and waterlogging, producing stagnogley podzols; a pattern found across western and upland UK, and western mainland Europe (Dimbleby, 1962;Duchaufour, 1982, pp. 112-121;Findlay et al., 1984;Gebhardt, 1993;Macphail, 1989). The identification of these paleosols demonstrates a considerable potential for the archaeological remains from the Early, Mid (and possibly later) Holocene to be contained, preserved and buried under the more recent peat accumulation (Gearey & Fyfe, 2016).
These analyses contextualize the presence of recent archaeological discoveries on Exmoor. At Hawkcombe Head, where Mesolithic pits and flints were preserved within and around a spring line (Gardiner, 2007), a brown earth paleosol can now be identified around the site. A Mesolithic heated pit (Sample 6) was recently excavated at Wintershead (Bray, 2017;Carey, 2017), which had been backfilled with brown earth soil. While the pit enclosure at Lanacombe is currently undated, based on morphology it is likely to be prehistoric; its pits [458] and [450] cut the brown earth and are buried by the podzolic peat, and defines the potential for the premire landscape of Exmoor to contain significant archaeological features. Likewise, the stone holes excavated by Gillings et al. (2010) had fill sequences containing fine brown silt, which can now be interpreted to have been derived from the brown earth paleosol that these stones were cut into, before the accumulation of peat.
The presence of these paleosols in various states of preservation is significant, given the relatively thin peat sequences of Exmoor. Relationships have been demonstrated between human activities destabilizing landscapes within fluvial catchments and the onset of alluviation. This has been used to define the Anthropocene through geomorphological change within alluvial systems, caused by anthropogenic disturbance of wider landscapes (Brown, Toms, Carey, & Rhodes, 2013). Likewise, the analysis of pollen data demonstrates increasing human interference and impact upon the environment from the Neolithic onwards, again demonstrating human drivers of ecosystem change in prehistory (Stephens et al., 2019). Data from the study of soils and paleosols during the Holocene can also play a key role in the definition of the Anthropocene, through consideration of the human contribution to evolution and transition of soils over time.
Human impacts on past environments and an assessment of the role of past human societies in terms of landscape evolution and soil degradation has significant potential to inform the definition and discussion of the Anthropocene.

| CONCLUSION
This study represents the first geoarchaeological analysis of a prepodzol brown earth soil on Exmoor, which in two locations was Both horizons contain very poorly sorted silts and fine sands (quartz, quartzite, feldspars, and micas), with medium and coarse sands and common gravel and small stones (≤22 mm), siltstones, fine sandstones, and shale, some of which are possibly subvertically imbricated. There are occasional weak organo-sesquioxidic staining becoming abundant, many thin burrows, becoming abundant upwards, and very abundant very thin organomineral excrements, with also many thin organomineral excrements are also present Context (453) lower is composed of lower subsoil B(s)/C(s) horizon material, where possibly relict imbricated shale rocks occur, and where an acid brown earth subsoil fine fabric is becoming very weakly affected by podzolization. Upwards (453) upper, remains of the likely relict Pleistocene stony head is present, alongside a more strongly podzolic and sesquioxidic pellety fine fabric (Bs horizon) (452) 21-11.5 Between 19 and 20 cm dominant dark reddish-brown humic and brown silt and sands, make up a band of sesquioxidic pellety humus (Bhs horizon) within the podzol. Between 11 and 19 cm the unit is heterogeneous with dominant weakly humic brown silt and sands and frequent broad chamber and channel fills with dark reddish-brown humic silt and sands. Gravel and small stones (>35 mm) occur. Rare root remains associated with ferruginization are present just below iron pan. There are many pellety organo-sesquioxidic polymorphic features. At 11.5 cm, there is a 2-4 mm thick amorphous iron pan, possibly cementing and replacing peaty sands, patches of very abundant thin burrows and chambers, and very abundant very thin organomineral excrements, with occasional thin organomineral excrements Mainly subsoil Bs horizon soil with mixing with humic pellety Bh horizon soil which becomes more dominant upwards, before being cemented into the iron pan (451) 11.5-2 A heterogeneous sample with dark reddish-brown organic silt and sands and pale greyish-brown minerogenic fine sandy silt loam, and frequent gray minerogenic silts and very fine sand. It is stony with very dominant gravel to small stone-size fragments (max 35 mm), with bleached rims. Rare thin in situ roots, showing weak iron staining, a rare trace of plant remains and fine charcoal (≤0.5 mm) also occur. The main depletion features are the bleached rims of rock fragments. There are abundant thin burrows, and very abundant thin and very thin organic excrements (which contain silt and fine sand) This is a stony dump of peaty topsoil mixed with leached E horizon and weathered sock and C horizon material. The presence of podzolic B horizon soil was not noted. The mixed topsoil material contains only a small trace of charcoal associated with colluvium before podzolization, suggesting clear human impacts within this landscape. It is noteworthy that while there was some research into paleosols on the upland landscapes of the southwest in the late 1970s and 1980s (principally Dartmoor), there has been a hiatus since. The analysis of these paleosols has significant potential to contribute to our understanding of human lifeways and economy in these remarkably rich upland archaeological landscapes.
The extensive preservation of prehistoric hut circles and field divisions across the uplands provides the possibility of combining the study of paleosols beneath these monuments with past environments and soil erosion, integrated with the analysis of paleoecological data, to understand past agricultural and site-specific environments. The identification of paleosols helps contextualize prehistoric finds and provide a linkage between localized pollen sequences and archaeological site-specific environments within these landscapes. With

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.