3.2.2 Vegetation and Woodland

Pollen can provide information on cultivated and wild plants along with woodland. Information on pollen is generally gathered now in most projects, although as shown in section 3.2.1, is rarely pulled together to provide a local or regional picture (but see Bunting and Farrell 2018). Although pollen is the most common material used to document and date vegetation changes, Birks (2003) argued for the need to validate this with other independent evidence such as macrofossils, which unlike wind-dispersed pollen are more likely to be local indicators. However, it is important to note that pine needles degrade in peat and that pine pollen can travel far from the parent tree. Dates obtained from pine pollen and macrofossils, both from the same context in Abernethy, Strathspey had a difference of around 150 years (Birks, 2003). Studies have shown that pollen percentages of just 1.5–3% total land pollen may indicate the local presence of pine (see discussion in Timpany et al 2020, 16).

Some pollen evidence may relate to the local situation while others might provide a more regional picture. The pollen collected from Loch Sionascaig in Wester Ross and the largest island in the loch, Eilean Mor, was intensively studied through pollen analysis, diatom analysis and sediment geochemistry. The Loch Sionascaig material is likely to have derived from a large source area and has been interpreted as a regional record. Pollen investigation of a bog on Eilean Mor provided a different pollen diagram from the loch; this provides a good picture of the development of peat (Birks 1993). The results of the research at Eilean Mor reflects the need for a multiple core approach.

The general picture of woodland colonisation, including the importance and widespread nature of hazel is noted in the National ScARF (Palaeolithic and Mesolithic section 3.2.2). This leads to a general predominance of birch, hazel and alder, with survival in places of oak and pine. However other studies allow for further nuances to be discerned. Evidence of past woodland can be determined from pollen, charcoal, and macrofossil data including from preserved seeds. Charcoal in particular can provide a wealth of information, not only for studying species which are often hidden in pollen analysis and dating, but also for understanding cutting techniques and insights into woodland management, including burning, possible coppicing and pollarding.

Richard Tipping’s overview (1994) that includes a wealth of information for the Highlands is based on over 40 sites, though much of the work was done before secure radiocarbon dating. However, he noted that very few pollen diagrams from the northern mainland were keyed to archaeological concerns. He cited the cases of An Druim and Loch Assynt in Sutherland and Loch of Winless in eastern Caithness, and questioned whether the focus on lake sites has given a misleading impression of human activity, or the lack thereof, in the record (Tipping 1994, 24). Since 1994 there have been a few other studies of this kind (see Map 3.2), but there are still undoubtedly gaps in our record

Map 3.2  Pollen studies with more than one radiocarbon date
Based on Tipping 1994: black numbers relate to sites. Green stars: selected studies since 1994. © Scott Timpany

Surviving woodland today may be less diverse than in prehistory, as pollen analysis shows that there was more variety even within some regions. Small areas of woodland in largely treeless landscapes can be missing in some pollen records but detected in others, suggesting that closer investigation is needed in many of these areas (Bunting and Farrell 2018). More evidence of waterlogged plant remains is needed to go alongside pollen in order to investigate the biodiversity.

More attention should be directed towards discerning whether one tree species dominated primary woodlands or whether mixing was the norm and if this changed over time. See for example, work at Glen Affric (Tipping 2003; Davies and Tipping 2004; Davies et al 2004; Tipping et al 2006); Migdale, Sutherland (Davies et al 2017) and the Garbh Allt catchment area near Golspie (Tipping and McCullagh 2003; Tipping et al 2008a; 2008b). The Highlands has good potential for the exploration of species diversity using pollen and macrofossil evidence.

Tipping (1994) has shown that there are general differences in the type of wood cover in areas in the Highlands. Central Sutherland and Caithness were predominantly colonised by pine, with different periods of decline and sometimes regeneration in different areas (Davies et al 2017). South of the Great Glen there was also pine, but the records show different patterns of clearance and recolonisation (Wilson et al 2011). To the west pine was not dominant, replaced by oak in places (Green and Edwards 2009). On Skye, birch and hazel were the main species, but more studies are needed to show if this interpretation is too generalised (Tipping 1994). There are other areas which need more work and more studies involving radiocarbon dating, including Easter Ross, areas along the Great Glen and Nairnshire. Even within these generalisations, however, studies have shown survival of woodland in microclimates (Davies et al 2017). Recent work investigating native woodland cover in certain areas of the Highlands is being used to inform replanting schemes (Davies 2011; Sybenga 2020; work in progress by Rob Wilson at St Andrews University); these projects are a good example of how commercial concerns can be allied to archaeological research.

Much of the work in the Highlands relates to woodland growing on peat, and further work is needed to see if there are differences for those growing on mineral soils, as was tackled in work at Glen Affric (Davies et al 2004; Tipping et al 2006) and the Garbh Allt catchment area near Golspie, Sutherland (Tipping and McCullagh 2003; Tipping et al 2008a; 2008b).

In the Highlands, there has also been attention directed towards the spread and demise of Scots pine (eg Birks 1972; 1975; 1993; Blackford et al 1992; Charman 1994; Tipping et al 2007b; 2008a: 2008b; Timpany 2008; Wilson et al 2011; Davies et al 2017). Early evidence for colonisation in the region survives around Loch Maree in Wester Ross from around 6500 BC. The source of the seeds is not clear, though they perhaps came from Ireland. Without oak or elm to compete against, pine dominated throughout much of the Highlands. However, the picture is complicated, as genetic differences between pines in different areas have been identified (Birks 1972; Tipping 2003a, 20–22). Pine then has a complex history in the northern Highlands.

While a number of studies suggest an abrupt decline around 4000 cal BP/2050 cal BC (Wilson et al 2011), this now is seen clearly as too simplistic (Tipping et al 2008a, 252; Davies et al 2017).  Detailed work at Loch Farlary, showed two declines: the first between 6200–5500 cal BP and 4250–3550 cal BC caused by rapid and frequent climatic shifts and the second from around 4200–3300 cal BP to around 2250–1350 cal BC was far more gradual. This suggests that the second decline at Loch Farlary was not linked to climate though the reasons for this decline are not entirely clear (Tipping et al 2008a). Elsewhere such as parts of Glen Affric, there appears to have been the fairly continuous growth of pine (Tipping et al 2006). Altogether, this shows a complex and locally sensitive situation in the Highlands, with dendrochronology of pine stumps holding some promise for future refinement (Wilson et al 2011). There is potential to utilise multi-proxy approaches including analysis of pollen and insect micro fauna, and dendrochronology to really explore pine decline across the region.

The interpretation of human action in woodland management is a difficult one (Tipping 1994; Bishop et al 2015, 67–68). The need for further work to try and determine whether climate or human activity is the main factor in woodland changes has been highlighted by the National ScARF. How long did ‘primary’ woodland last in these areas of the Highlands? What would these ‘primary’ woodlands comprise of (ScARF Bronze Age section 3.1)?

A pollen modelling study in Coigach, Wester Ross, was pioneered to try and assess what pollen in the archaeological record can tells archaeologists when by also focussing on modern pollen rates and dispersal, and issues involving climate, hydrology, topography and landscape. The effect of woodland management practices on the pollen records are also beginning to be investigated though modelling; this is an area of investigation important for the Highlands (Bunting and Farrell 2018).