The climate and biomes of Europe at 6000 yr BP: comparison of model simulations and pollen-based reconstructions

Prentice, I. C., Harrison, S. P. ORCID: https://orcid.org/0000-0001-5687-1903, Jolly, D. and Guiot, J. (1998) The climate and biomes of Europe at 6000 yr BP: comparison of model simulations and pollen-based reconstructions. Quaternary Science Reviews, 17 (6-7). pp. 659-668. ISSN 0277-3791 doi: 10.1016/S0277-3791(98)00016-X

Abstract/Summary

14C-dated pollen and lake-level data from Europe are used to assess the spatial patterns of climate change between 6000 yr BP and present, as simulated by the NCAR CCM1 (National Center for Atmospheric Research, Community Climate Model, version 1) in response to the change in the Earth’s orbital parameters during this perod. First, reconstructed 6000 yr BP values of bioclimate variables obtained from pollen and lake-level data with the constrained-analogue technique are compared with simulated values. Then a 6000 yr BP biome map obtained from pollen data with an objective biome reconstruction (biomization) technique is compared with BIOME model results derived from the same simulation. Data and simulations agree in some features: warmer-than-present growing seasons in N and C Europe allowed forests to extend further north and to higher elevations than today, and warmer winters in C and E Europe prevented boreal conifers from spreading west. More generally, however, the agreement is poor. Predominantly deciduous forest types in Fennoscandia imply warmer winters than the model allows. The model fails to simulate winters cold enough, or summers wet enough, to allow temperate deciduous forests their former extended distribution in S Europe, and it incorrectly simulates a much expanded area of steppe vegetation in SE Europe. Similar errors have also been noted in numerous 6000 yr BP simulations with prescribed modern sea surface temperatures. These errors are evidently not resolved by the inclusion of interactive sea-surface conditions in the CCM1. Accurate representation of mid-Holocene climates in Europe may require the inclusion of dynamical ocean–atmosphere and/or vegetation–atmosphere interactions that most palaeoclimate model simulations have so far disregarded.