A word of caution when planning forest management using projections of tree species range shifts

In this note we raise our concerns about the use of climate envelope models as a basis for forest planning under climate change. Such models assume constant relationships among tree species presence, abundance or growth rates and climatic variables, and that these can be transferred from their current distribution areas to areas that are predicted to have a similar future climate. Climate is an important determinant of tree species distributions, but its effects are mediated through soils, competition from other plant species, herbivores, diseases, insects and fire. This complexity should be addressed when making predictions about plant species distribution changes. If forecasts based only on climate are accepted uncritically and become the basis for forest policy and practice, there could be important consequences for the success of forest management. We illustrate the issue with the historical response of tree growth to climate variability for three conifer species along an altitudinal gradient in southern interior British Columbia. The growth–climate relationships differ not only among species but also between ecological zones, which implies that the different combinations of tree species and site will react differently to the same change in climate. All things considered, caution is needed when developing management plans using predicted future tree distributions based only on current/past tree/climate relationships. Key words: climate-envelope models, climate change, species distributions, dendroclimatology