3.2 Vegetational and associated environmental changes during Late Devensian and early Holocene times

Vegetation history during the periods covered in this document is most readily reconstructed from pollen analysis (palynology), supplemented by the evidence from other indicators such as plant macrofossils, insects and sedimentology. This survey makes no pretence at exhaustiveness and the associated literature is vast. Reviews of vegetational history, often with a specific focus, may be found in Edwards and Ralston (1984), Tipping (1995; 2004), Edwards and Whittington (1997a, b; 2000; 2001), and Walker and Lowe (1997). Area-specific contributions include Keatinge and Dickson (1979), Birks and Williams (1983), Edwards and McIntosh (1988), Edwards (2000), Edwards et al. (2000b, 2005), Tipping and Milburn (2000), Tipping (2008), and Green and Edwards (2009). The techniques of palynology are discussed exhaustively in Dimbleby (1985), Faegri and Iversen (1989), and Moore et al. (1991). For the evidence from plant macrofossils, readers are directed to Godwin (1975) and Dickson and Dickson (2000); climate change events per se are summarised by Tipping (see Table 1).

3.1 Climate changes in Scotland from the Last Glacial Maximum c. 16000 yrs BP to c. 6000 BP

The current climato-stratigraphic framework or stratotype for climate change in the Late Devensian Epoch, intended to be applicable across the North Atlantic region, is described in Lowe et al. (2008), following re-calibration of the annually resolved NGRIP δ18O ice-core record by Rasmussen et al. 2008.  The table below presents the ages of successive climatic events, numbered from the base of the Holocene.  Ages are defined from the NGRIP record, given by Rasmussen et al. 2008 as GICC05 age ka b2k, or years before AD2000 in the NGRIP record. There is expected to be a close relation between these ages and, for instance, calibrated 14C assays (cal BP), but the imprecision of calibrations makes this not possible to establish.  GS events are colder, stadial events: GI events are milder interstadial events.

Table 1: Table outlining the ages of successive climatic events, numbered from the base of the Holocene.
Event Onset and Termination (GICC05 age b2k) Duration (years)
Holocene 11703 to present 11203
GS-1 12896 to 11703 1193
GI-1a 13099 to 12896 203
GI-1b 13311 to 13099 212
GI-1c 13954 to 13311 843
GI-1d 14075 to 13954 121
GI-1e 14692 to 14075 617
GS-2a Not defined to 14692  

Greenland Stadial 1 (GS-1) is roughly equivalent to the Younger Dryas Stadial; Greenland Interstadial 1 (GI-1) lasting 1996 years is roughly equivalent to the Windermere Interstadial.  GI-1d and GI-1b are colder phases within GI-1; GI-1d may be equated with the ‘Older Dryas’ of continental NW Europe.

Temperature variations in Scotland prior to the Holocene are best defined from the Whitrig Bog chironomid record in SE Scotland (Brooks and Birks 2000), but this sequence is not dated.  From biostratigraphic and tephrostratigraphic correlation (Turney et al. 2007), chironomid assemblages suggest that mean July temperatures were lower than c. 7.5°C before the Windermere Interstadial, perhaps before c. 14300 GICC05 age ka b2k, and reached c. 12°C after this.  Coleopteran data (Coope 1987) suggest temperatures of c. 18°C. Temperatures decline throughout the Windermere Interstadial in all records to around 11°C.  Evidence from chironomids suggests three falls in mean July temperature of between 0.5 and 3 °C within the Windermere Interstadial, perhaps equated with events GI-1d and GI-1d.  Perhaps around 12896 GICC05 age ka b2k, mean July temperatures plummeted to below c. 7.5°C.  Within the Younger Dryas, mean July temperatures rose to c. 9°C.  Assuming precise correlation with the NGRIP record, the Younger Dryas succeeded to the Holocene Epoch at around 11703 GICC05 age ka b2k.  In Greenland this occurred in a matter of decades (Taylor et al. 1993; Alley 2000).

Changes in precipitation are very poorly understood from Scottish Lateglacial sequences.  The greater climatic continentality of The Netherlands, and a heightened sensitivity to precipitation of sediment sequences on well-drained sand allows the suggestion of a wet early Windermere Interstadial, but increasing aridity to the Older Dryas (≈GI-1d: 14075 to 13954 GICC05 age ka b2k) (Walker et al. 1994).  Effective precipitation (precipitation – evapotranspiration) was variable after this until the Younger Dryas, which was again arid.

The earliest Holocene climatic amelioration was interrupted by a series of abrupt, hemispheric or global climatic events.  The establishment of Betula (birch) woodland in the Netherlands from c. 11500 cal BP (Friesland Phase) was interrupted around c. 11400 cal BP by a dry, continental climate, the Rammelbeek Phase, correlated with the Preboreal Oscillation in the NGRIP record.  Effective precipitation increased after c. 11250 cal BP (9300 cal BC; Bos et al. 2007), with the re-establishment of the North Atlantic Current (Andresen et al. 2007).  Dense woodland was not established in The Netherlands until c. 10730 cal BP (8780 cal BC; Bos et al. 2007), at what used to be the beginning of the Boreal period.  Further short-lived deteriorations in climate occurred at c. 10300 (8350) and c. 9500 cal BP (7550 cal BC), seen in some but not all proxies (Hoek and Bos 2007 and references therein).  The major early Holocene climatic reversal, the 8.2 ka event (Alley et al. 1997), was small in comparison with Lateglacial oscillations but had widespread, hemispheric impacts and involved a temperature depression of 2–3°C (Klitgaard-Kristensen et al. 1998).  North-west Europe was markedly more arid.  Stager and Mayewski (1997) and Debret et al. (2009) argue that northern hemispheric atmospheric circulation before c. 8200 cal BP (6250 cal BC) was probably markedly different to today because of the persistence of a Laurentide ice cover; recognisable ‘Holocene’ climatic patterns may have commenced only at this time.

No significant climatic reversals are recorded in NW European proxies until c. 6000 cal BP (Mayewski et al.2004).  The period 5400 to 4000 BC was 1–2°C warmer than present in NW Europe (Davis et al. 2003). However, the period of hemispheric abrupt climate change that would later be associated with Neolithic archaeological events actually commenced in the preceeding Mesolithic, from c. 7000 cal BP (5050 cal BC).  

Table of chronostratigraphic and nomenclatural comparisons. Ice core ages (ice core years before AD 2000) and events from Lowe et al. (2008) and Walker et al. (2009). IACP = Intra Allerød Cold Period.