Anne Crone
As well as providing precise calendar dates dendrochronology also has the potential to offer proxy evidence on the woodland resource and its exploitation. In this case study the power of dendrochronology to animate the history of a settlement is presented.
The wetland settlement at Black Loch of Myrton, Dumfries & Galloway was built on a damp peaty islet projecting out into what would have once been a small shallow loch. Submerged and relatively undisturbed for over two millennia, agricultural drainage and increasingly climate change were beginning to impact on the water levels which protected the site, leaving areas in which the organic remains were beginning to dry out and decay. Supported by a grant from Historic Environment Scotland, AOC Archaeology Group carried out excavations there between 2013 and 2019. Excavations revealed a settlement comprising a group of roundhouses within a palisaded enclosure, the houses lying on either side of a log trackway which linked the settlement with the shore. Preservation in the southern half of the site was excellent, to the extent that the plank and wickerwork walls of several of the roundhouses survived to heights of 0.4 m above the original ground surface, while sections of the log trackway remained intact.
Dendrochronological analysis of the wood used to construct the settlement has, together with an extensive radiocarbon-dating programme, enabled us to develop a detailed chronology for its development, from initial construction to abandonment. Analysis of the oak has provided the precise calendrical framework for the site while analysis of the alder, hazel and ash has provided additional information on the relationship between structures and their durations. There were three main episodes of activity on the settlement.
Episode 1
The primary settlement was constructed over a period of two years. Felling of timber began in 437 BC. The trackway into the settlement began to be laid in that year, work progressing into the interior with the construction of a palisade of alder posts with oak gateposts in the winter/spring of 436/435 BC. ST2 was constructed in the spring/summer of 435 BC and ST1 was built in the winter/spring of 435/434 BC. Analysis of the ash from both structures suggests they were both built in the same year so ST1 was probably also built in 435BC. Analysis of hazel stakes also suggests that ST2 and ST5 were built in the same year.
There is no dendrochronological evidence in ST1 and ST2 for anything more than a single phase of construction of the superstructure and modelling of the radiocarbon dates suggest that ST1 and ST2 were in use for no more than 40 years, so they were probably abandoned circa 400 BC.
Episode 2
Oak was rarely used in the structures of this Episode which means that it could not be precisely dated using dendrochronology alone. The dendro date of 436/435 BC for the Episode 1 palisade provides a clear terminus post quem for some of the Episode 2 houses which were built over the palisade and the radiocarbon dates suggest that they were built circa 400 BC, ie almost immediately after the Episode 1 buildings were abandoned. A small number of decayed oak planks on the trackway suggests building activity on the trackway sometime after 410 BC and probably in the first quarter of the 4th century BC. The correlation between a replacement post of ash in ST3 and an ash post alongside the Episode 1 trackway in the entrance suggests that a refurbishment phase in ST3 took place some 66 years after the trackway construction. As oak timbers used in the trackway were felled between 437 BC and 435 BC this would place the refurbishment phase between 371 BC and 369 BC. Modelling of radiocarbon dates suggests that ST3 fell out of use between 375-310 cal BC, thus corroborating the dendro date.
The only Episode 2 structure that produced dendrochronological results is ST3. Analysis of the alder has suggested that the first two phases of activity in the house occurred within the same year, while analysis of the hazel stakes suggests there may have been as little as 16 years between the construction of the outer wall line and its final refurbishment.
Episode 3
The only archaeological evidence for this episode consists of the defences built around the southern perimeter of the island. Analysis of the oak timbers used in their construction shows frequent rebuilding of the settlement perimeter over 55 years, from 278 – 276 BC to 223 BC. At least three distinct building events have been identified. Between 278 to 276 BC a barrier of posts was constructed on either side of the trackway in the entrance area onto the settlement. In 243 BC an oak plank palisade was constructed on the very edge of the islet, forming an impressive façade on either side of the entrance way.
There was further building work on the southern perimeter of the settlement in 223 BC, represented by a single post. If we assume that occupation continued after this building event, and allow a generation at least for that occupation, then the settlement was probably finally abandoned in the early 2nd century BC.
Iron Age woodlands
Throughout the occupation of the settlement hazel, alder and oak were the favoured species for building and for fuel.
Two woodland types are indicated by the age and size of the oak timbers. The posts were from relatively young, medium-sized roundwood trunks; although their girth does not change greatly from Episode 1 (194 mm av diam) to Episode 3 (207 mm av diam) the average age of the trees is generally younger by Episode 3, reducing from 97 yrs to 67 years. The oak posts could have come from an open woodland of maiden trees from which timber had been regularly extracted, hence the relatively young ages of the trees.
In contrast, the oaks converted to make the vertical planks used in Episodes 1 and 3 were significantly larger and older trees. Some of the parent trees will have been around 500 years of age, with estimated diameters of between 0.6 and 0.7 m. These trees represent a very ancient woodland which must have been protected from exploitation, allowing the trees to grow to great ages. Nearly two centuries passed between their felling in 437-5 BC in Episode 1 to their felling in 243 BC in Episode 3 and yet there is very little change in the range and average age of the ancient oaks over that period suggesting that this woodland was managed to ensure that there was a continuous supply of oaks of this size. This, in turn implies that the woodland was owned, or controlled. It seems probable that these oak woodlands were not that close to the settlement because very little oak was used as fuel; the trimmings off the trunk would have provided a useful source of fuel unless it required an unequal amount of energy to transport it to the settlement.
There is very little change in the age and size structure of the hazel assemblage from Episode 1 to Episode 2, the main difference being that some older wood was cropped in Episode 1. Again, this suggests that the hazel woodland was being managed to supply the wood in the type and quantity that the settlement needed.
Some 77% of the analysed alder came from Episode 1 and all the timbers over 70 years of age came from this Episode. There may have been areas of ancient alder carr nearby supporting the maiden trees which were able to provide the settlement with the large, straight-grained logs used in its construction. In Episode 2 the alders used in ST3 are only 31 years of age on average and much smaller, at 95 mm in average diameter. This suggests that the carr had been exhausted in building the primary settlement, leaving only smaller trees available for use in Episode 2. The use of alder as a fuel also decreased significantly, from 51% in Episode 1 to only 22% in Episode 2, suggesting that by then local sources of alder were not so easily available any longer.
The extensive dendro programme at Black Loch of Myrton has revealed a tempo of activity which allows us to construct narratives at the scale of human lifetimes. By recording the species, diameter, age and growth-patterns we have been able to reconstruct the woodland resources available to the occupants and to detect changes to those resources over time, which probably reflect management practices and land ownership.