Get access

Landscape planning for the future: using fossil records to independently validate bioclimatic envelope models for economically valuable tree species in Europe

Authors

  • Marc Macias-Fauria,

    Corresponding author
    • Long-term Ecology Laboratory, Biodiversity Institute, Department of Zoology, University of Oxford, Oxford, UK
    Search for more papers by this author
  • Katherine J. Willis

    1. Long-term Ecology Laboratory, Biodiversity Institute, Department of Zoology, University of Oxford, Oxford, UK
    2. Department of Biology, University of Bergen, Bergen, Norway
    Search for more papers by this author

  • Editor: Solomon Dobrowski

Correspondence: Marc Macias-Fauria, Long-term Ecology Laboratory, Biodiversity Institute, Department of Zoology, University of Oxford, Tinbergen Building, South Parks Road, Oxford OX1 3PS, UK.

E-mail: marc.maciasfauria@zoo.ox.ac.uk

Abstract

Aim

Bioclimatic envelope models (BEMs) for seven economically important tree species in Europe were independently validated using a hindcasting approach and fossil pollen records spanning the last 1000 years, including the Medieval Warm Period (MWP), the Little Ice Age (LIA) and the 20th century (PRES). The aim was to determine the accuracy of combining BEMs and palaeoecological data to predict continental-scale changes in distribution, and the availability of fossil data to hindcast economically important species.

Location

Europe, 11 °W–33 °E, 33–72 °N.

Methods

Eight types of BEMs were implemented in this study, covering most state-of-the-art modelling techniques. Present and palaeoclimatic data were obtained from the atmosphere–ocean global circulation model ECHO-G. The last millennium was divided into three climatically distinct periods: the MWP (ad 900–1300), the LIA (ad 1600–1850) and PRES (ad 1900–2000). Models were calibrated for each period and validated with climatic and pollen data from the remaining periods. Successfully validated models were projected onto a 1° European grid.

Results

BEMs were successfully validated with independent data. The average area under the curve statistic showed strong model performance, indicating: (1) the strength of BEMs for modelling current and future distributions and, (2) the potential of fossil pollen records to undertake these approaches. European-scale 1°- gridded maps of probability of occurrence largely agreed with observed 20th-century distributions for most taxa, allowing the construction of past modelled species distributions.

Main conclusions

Results suggest a high potential for BEMs to be used to model future species distributions, and highlight the importance of palaeoecological data to independently validate these models, taking into account the scales at which these data operate. Although valid, BEMs showed poorer performance with heavily managed species and/or those growing in heterogeneous terrain or with discontinuous distributions. The last millennium in Europe was characterized by an increase in woody crop species and a decline of forest species, suggesting increasing land use by humans.

Ancillary