Abstract :
Past simulation studies using a variety of models have generally agreed that climatic warming could have adverse effects on forests, including large-scale diebacks in some regions and drastic range shrinkages of many species. These effects should be most evident at biome transition zones. Other studies have pointed out, however, that past models have used a parabolic temperature response function that is based on geographic range limits rather than functional responses or data and that this parabolic model could exaggerate dieback effects. A new model is proposed for growing degree-days temperature response, which is asymptotic rather than parabolic. In this new model, tree height growth rate increases and then levels off with increasing growing degree-days. Species from more southern regions have a higher asymptote. It is shown that this model can be derived from the integration of a parabolic growth response to temperature over a year-long sinusoidal temperature regime. The SORTIE forest simulation model was modified to incorporate this response function. An ecotonal region was simulated under a warming scenario. The traditional parabolic temperature response model produced a wide zone of dieback following warming. In contrast, the new asymptotic response function produced no dieback and a stable ecotone that migrated north at <100 m/100 years.
