The overall interpretation of the palaeoenvironment and land use at the three sites studied and at other comparable sites in the region is presented as a timeline in Table 18.
The earliest palaeoenvironmental evidence from the three early field systems is from the charcoal records and soil pollen records which provide some limited evidence for the nature of woodland and possible human activity prior to the time the early fields were constructed. The charcoal is basically residual and represents fire events, possibly associated with clearance activity and attempts at management of the vegetation and landscape prior to construction of the field systems. The earliest evidence is from a possible post-hole at Muriau Gwyddelod where hazel charcoal is dated to 7190–7060 and 7300–7220 cal BC. Here the assemblage suggests oak, hazel and alder woodland during the Mesolithic, and is in agreement with pollen evidence from a pollen site at Moel y Gerddi (Chambers and Price 1985; 1988). Previously, on the basis of charcoal in the peat stratigraphy at Moel y Gerddi, it had been suggested that there might be Mesolithic activity in the area, though there was a lack of archaeological evidence to support this, and that the early establishment of alder carr might have been aided by human activity (Chambers and Price 1985, Chambers and Price 1988).
The continued presence of oak and alder woodland in the Muriau Gwyddelod area is suggested by a date of 3250–3100 and 3360–3260 cal BC from the alder charcoal from the buried soil (15), which suggests possible clearance activity during the Middle Neolithic. Further evidence for human impact in the area during the Neolithic is provided in the pollen record from Moel y Gerddi (Chambers and Price 1988). At Muriau Gwyddelod hazel charcoal was recorded from the buried soil (14) and the pollen evidence from beneath the field bank suggests oak-hazel woodland, with some alder woodland, and an increasingly open pastoral landscape prior to establishment of the bank, though with some arable cultivation in the area immediately before or contemporary with construction of the bank There is also pollen evidence from the mire at Moel y Gerddi and from the late prehistoric enclosures of Erw-wen and Moel y Gerddi for cultivation, as well as charred cereal, both wheat and barley, from Erw-wen.
Charcoal also provides the earliest evidence from the Cwm Cilio and Braich y Gornel areas. Oak charcoal, for a fire event dating to 5210–4990 cal BC at Cwm Cilio might represent late Mesolithic interference with woodland in the area rather than a natural fire, while a further fire episode dated to 1910–1740 cal BC might indicate clearance of woodland, including hazel, during the Early Bronze Age. A date of 1310–1050 cal BC from hazel charcoal at Braich y Gornel possibly suggests that the clearance of oak and hazel woodland by the use of fire continued into the Middle Bronze Age. Although the pollen records from the Cwm Cilio and Braich y Gornel pollen sites do not extend back to cover these periods, the palaeoenvironmental record from Cefn Graeanog (Chambers 1998), lying slightly to the north of the two sites, extends back to c.6500 cal BC.
The record from Cefn Graeanog provides some circumstantial evidence for human activity during the Later Mesolithic, including an expansion in sorrel which might signify openings in the woodland, but there is no microscopic charcoal evidence from the site which might have indicated fire activity during that period. There is, however, macroscopic charcoal evidence in the stratigraphy as well as pollen evidence for interference with the woodland from the Neolithic onwards, with evidence for a more marked impact on woodland during the Middle Bronze Age.
The evidence from the pollen cores from Cwm Cilio and Braich y Gornel probably post-dates the charcoal evidence, although comparison of the two records suggests that the record from Braich y Gornel may begin slightly earlier, possibly during the mid- to late Bronze Age, with that from Cwm Cilio possibly commencing during the late Bronze Age or Iron Age. The pollen records from the two sites display a similar sequence of vegetation episodes suggesting they share a common regional environmental history, although local differences can be detected, probably reflecting their different locations. By the time the records commence they suggest that the landscape was already substantially opening up, which is consistent with the charcoal evidence from the sites which indicates earlier fire activity in the area, possibly associated with clearance. Although some woodland persisted, largely alder but with hazel, oak and birch in the Cwm Cilio area and alder and hazel in the Braich y Gornel area, suggesting some local woodland variation within the region, the dominant land use was pastoral farming. Dates of cal AD 0–130 and cal AD 20–130 from Cwm Cilio and Braich y Gornel, respectively, date the beginning of a marked decline in woodland in both areas, following a brief period of regeneration. There is also tentative evidence for cereal cultivation around this time which may relate to the roundhouse settlements and field systems.
There is further palaeoenvironmental evidence in the local region from Cefn Graeanog (Chambers 1998) around this period but, as at Cwm Cilio and Braich y Gornel, there is some uncertainty over the dating evidence. It is suggested that there was a major episode of woodland clearance affecting alder and oak c.400–200 cal BC which related to an agricultural phase followed by the first building phase at Cefn Graeanog II. This period of Iron Age woodland clearance and agricultural activity, largely pastoral, recognised at Cefn Graeanog may broadly correspond with the evidence for clearance and pastoral land use from Cwm Cilio and Braich y Gornel and was perhaps associated with some of the roundhouses. This was, in turn, succeeded by a phase of hazel and birch secondary woodland which lasted until c.cal AD 50 and coincided with a period of abandonment at the site, though other settlements on the Graeanog ridge continued to be occupied. The recovery in woodland recognised at Cefn Graeanog, although birch and hazel rather than mainly alder as at Cwm Cilio and alder and hazel at Braich y Gornel, seems to represent a similar regeneration event at all three sites and the dates of cal AD 0–130 from Cwm Cilio and cal AD 20–130 from Braich y Gornel are broadly in keeping with that estimated for Cefn Graeanog.
After this, major interference is considered to have occurred in the landscape at Cefn Graeanog, probably related to the main period of occupation at Cefn Graeanog II as well as other hut groups on the ridge. However, dates of 1035±45 BP (CAR-69) and 1130±45 BP (CAR-72) relating to this clearance episode and the following regeneration episode appear to be much too young. This may have been as a result of the roots of later vegetation rather than disturbance of the peat record because an older date of 1225±40 BP (CAR-68) occurs higher in the stratigraphy, although there was evidence for disturbance in an adjacent peat section (Chambers 1998). The dates from Cwm Cilio and Braich y Gornel for the preceding regeneration phase also seem to support the suggestion that these dates are too young and, as mentioned above, these dates are consistent with the estimated date for the end of the regeneration phase at Cefn Graeanog, which was based on the date of 1225±40 BP (CAR-68) being correct, rather than the rejected dates. All three sites show a substantial fall in arboreal pollen values following the regeneration episode, suggesting considerable woodland clearance in the region and an expansion in agricultural activity, which, on the basis of the previous argument, probably occurred in the Late Iron Age/Roman period and related to the settlements and field systems in the three areas. The strongest evidence for arable cultivation is at Cefn Graeanog and is probably related to the Romano-British settlement at Cefn Graeanog II.
All three sites indicate pastoral activity and both wetland and dryland pasture. Species such as bog asphodel, indicating base poor soils, occur at all three sites. This phase of farming activity continues throughout the Roman period and is followed by some woodland regeneration in the region, though the nature and timing of it seems to vary. At Cefn Graeanog, birch, alder, oak and hazel increase before a renewed decline c.cal AD 700–900 and, similarly, at Cwm Cilio a brief recovery in alder, hazel and birch, followed by a decline, had taken place well before cal AD 1030–1210. At Braich y Gornel there is a slight recovery in hazel but the beginning of this is dated to cal AD 1020–1050 or cal AD 1080–1130, if the date is accepted as correct and not contaminated, as at Cefn Graeanog. If accepted as correct then the record is very compressed. Another possibility is that high levels of microscopic, as well as macroscopic, charcoal in the peat at Braich y Gornel represent burning of the bog surface and possibly disturbance of the pollen record. Similarly, at Cwm Cilio high levels of microscopic, as well as macroscopic, charcoal may indicate interference with the bog surface and there was some evidence of peat cutting and drainage at Cwm Cilio, although this was not immediately apparent at Braich y Gornel. This activity may relate to medieval and/or post-medieval occupation in the areas.
An open landscape is suggested for all three sites for the remainder of the medieval and post-medieval periods apart from a brief increase in alder and hazel woodland at Braich y Gornel during the post-medieval period. The records do, however, allow some variation in agricultural land use in the region to be discerned, with possibly the greatest amount of cultivation taking place in the area around Cefn Graeanog, followed by Cwm Cilio and only limited cultivation, if any, taking place in the Braich y Gornel area.
Finally, in terms of soil development at the three sites, the evidence suggests that a colluvial plough soil had been present before pasture development at Braich y Gornel; clayey subsoils formed in glacial drift had been eroded down slope. At Muriau Gwyddelod, Fronhill, acidifying buried pasture soils were found, together with evidence of earlier clearance and the original earthworm-worked brown earth soil. A history of acid brown earth subsoil erosion and the deposition of eroded subsoil clasts down slope were recorded at Cwm Cilio, as probable evidence of arable activity before an ensuing land use change to pasture. It can be suggested that the soils are rather anomalous compared to the current and dominant soil types at all three localities. At both Braich y Gornel and Cwm Cilio, the buried soils were less acidic and better drained compared to the local cover. At Muriau Gwyddelod, Fronhill, the ancient soils have been acidifying since clearance and its earliest use as pasture.
Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.
Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.