• Global warming;
  • plant migration;
  • biomes;
  • greenhouse effect;
  • biodiversity


Greenhouse-induced warming and resulting shifts in climatic zones may exceed the migration capabilities of some species. We used fourteen combinations of General Circulation Models (GCMs) and Global Vegetation Models (GVMs) to investigate possible migration rates required under CO2-doubled climatic forcing.




Migration distances were calculated between grid cells of future biome type x and nearest same-biome-type cells in the current climate. In `base-case' calculations, we assumed that 2 × CO2 climate forcing would occur in 100 years, we used ten biome types and we measured migration distances as straight-line distances ignoring water barriers and human development. In sensitivity analyses, we investigated different time periods of 2 × CO2 climate forcing, more narrowly defined biomes and barriers because of water bodies and human development.


In the base-case calculations, average migration rates varied significantly according to the GVM used (BIOME3 vs. MAPSS), the age of the GCM (older- vs. newer-generation GCMs), and whether or not GCMs included sulphate cooling or CO2 fertilization effects. However, high migration rates (≥ 1000 m year−1) were relatively common in all models, consisting on average of 17% grid cells for BIOME3 and 21% for MAPSS. Migration rates were much higher in boreal and temperate biomes than in tropical biomes. Doubling of the time period of 2 × CO2 forcing reduced these areas of high migration rates to c. 12% of grid cells for both BIOME3 and MAPSS. However, to obtain migration rates in the Boreal biome that were similar in magnitude to those observed for spruce when it followed the retreating North American Glacier, a radical increase in the period of warming was required, from 100 to >1000 years. A reduction in biome area by an order of magnitude increased migration rates by one to three orders of magnitude, depending on the GVM. Large water bodies and human development had regionally important effects in increasing migration rates.

Main conclusions

In conclusion, evidence from coupled GCMs and GVMs suggests that global warming may require migration rates much faster than those observed during post-glacial times and hence has the potential to reduce biodiversity by selecting for highly mobile and opportunistic species. Several poorly understood factors that are expected to influence the magnitude of any such reduction are discussed, including intrinsic migrational capabilities, barriers to migration, the role of outlier populations in increasing migration rates, the role of climate in setting range limits and variation in species range sizes.