We present pollen-based reconstructions of the spatio-temporal dynamics of northern European regional vegetation abundance through the Holocene. We apply the Regional Estimates of VEgetation Abundance from Large Sites (REVEALS) model using fossil pollen records from eighteen sites within five modern biomes in the region. The eighteen sites are classified into four time-trajectory types on the basis of principal components analysis of both the REVEALS-based vegetation estimates (RVs) and the pollen percentage (PPs). The four trajectory types are more clearly separated for RVs than PPs. Further, the timing of major Holocene shifts, rates of compositional change, and diversity indices (turnover and evenness) differ between RVs and PPs. The differences are due to the reduction by REVEALS of biases in fossil pollen assemblages caused by different basin size, and inter-taxonomic differences in pollen productivity and dispersal properties. For example, in comparison to the PPs, the RVs show an earlier increase in Corylus and Ulmus in the early-Holocene and a more pronounced increase in grassland and deforested areas since the mid-Holocene. The results suggest that the influence of deforestation and agricultural activities on plant composition and abundance from Neolithic times was stronger than previously inferred from PPs. Relative to PPs, RVs show a more rapid compositional change, a largest decrease in turnover, and less variable evenness in most of northern Europe since 5200calyrBP. All these changes are primarily related to the strong impact of human activities on the vegetation. This study demonstrates that RV-based estimates of diversity indices, timing of shifts, and rates of change in reconstructed vegetation provide new insights into the timing and magnitude of major human disturbance on Holocene regional vegetation, features that are critical in the assessment of human impact on vegetation, land-cover, biodiversity, and climate in the past.
This work is a part of the LANDCLIM (LAND cover – CLIMate interactions in NW Europe during the Holocene) project and research network coordinated by M.-J. Gaillard and sponsored by the Swedish (VR) and Nordic (NordForsk) Research Councils 2009–2011. LANDCLIM is also a contribution to the Strategic Research Area MERGE (ModElling the Regional and Global Earth system; http://www.cec.lu.se/research/merge) coordinated by M. Rummukainen (Lund University, Sweden) and sponsored by the Swedish Research Council (VR) 2010–2014, and to the IGBP-PAGES-Focus 4 programme 2009–2014 (http://www.pages-igbp.org/science/foci/focus-4). We thank all members of the LANDCLIM network who contributed pollen data and for useful discussions and advices, as well as J.S. Carrión and the anonymous reviewers for valuable comments and suggestions. Other funding sources include the Faculty of Health and Life Sciences of Linnaeus University (L. Marquer, M.-J. Gaillard, A.-K. Trondman, S. Sugita), the French CNRS (F. Mazier) and the King Carl XVI Gustaf's Foundation for Environmental Sciences in Sweden (S. Sugita.). H.J.B. Birks acknowledges the assistance of C.R. Jenks. This is a contribution from the Bjerknes Centre for Climate Research (H.J.B. Birks and A.E. Bjune).
- Human impact
- Northern Europe
- Quantitative regional plant abundance
- Rate of compositional change
- REVEALS (Regional Estimates of VEgetation Abundance from Large Sites) model
- Vegetation diversity indices
FingerprintDive into the research topics of 'Holocene changes in vegetation composition in northern Europe: Why quantitative pollen-based vegetation reconstructions matter'. Together they form a unique fingerprint.