, 2012). Much uncertainty is also due to the unknown future trajectories of greenhouse gas emissions in the longer term, as these will depend on technological developments that increase or decrease emissions (IPCC, 2011). For more immediate future scenarios, however, the variation amongst models is small; for México, for the decade centered on the year 2030, for example, it is only about ±0.2 °C of mean annual temperature (Sáenz-Romero et

al., 2010). Another difficulty in modelling is that the current distributions of tree species, which form the basic input data for determining likely future distributions, are often not well known (McLachlan et al., 2007 and Rehfeldt and Jaquish, ABT 199 2010), especially in the tropics, where sometimes complex topographies and high biodiversity paradoxically make accurate predictions even more urgent. In the light of uncertainties in modelling, the United Kingdom’s Forestry Commission (2011) considers risk minimisation as the best

approach, by maintaining existing genetic variation, promoting migration, encouraging natural regeneration and supporting provenance mixing in plantations (Hubert and Cottrell, 2007). As already noted (see Section 4.2), interventions that involve moving tree species into entirely new areas is hotly debated because of potential disturbances to indigenous flora and fauna. There are also numerous commercial forestry examples where the introduction of ill-adapted genetic resources has resulted in massive production failures. Selleckchem PR 171 For example, 30,000 ha of Pinus pinaster Aiton plantations were destroyed in the Landes region of France during the winter of 1984 to 1985 following the introduction of non-frost-resistant material from the Iberian Peninsula ( Timbal et al., 2005). Careful thought to all environmental

factors should therefore be given before climate-related assisted migrations are undertaken. In mountain regions, upwards associated translocations may not be an option if populations are already at or near the summit (translocation must then be to different mountains), new or if edaphic conditions are unsuitable ( Lauer, 1973). Certainly, the establishment of viable populations at extremely high altitudes would be very challenging ( Sáenz-Romero et al., 2010 and Sáenz-Romero et al., 2012). Another challenge to assisted migration that is specific to long-living perennials is that, where climate is changing quickly, large differences in conditions may be observed over an individual tree’s lifespan. To find species or genotypes well adapted to conditions at establishment and at productive maturity (e.g., for some species, perhaps a century later) may therefore be difficult. In order to achieve a proper balance, the interval to production/maturity needs to be considered, and multiple stepped translocations over time may be required (Soto-Correa et al., 2012).