Physiology-based phenology models for forest tree species in Germany.

Models of phenology are needed for the projection of effects of a changing climate on, for example, forest production, species competition, vegetation-atmosphere feedback and public health. A new phenology model for deciduous tree bud burst is developed and parameters are determined for a wide geographical range (Germany) and several forest tree species. The new model is based on considerations of simple interactions between inhibitory and promotory agents that are assumed to control the developmental status of a plant. Several alternative model structures were formulated emphasizing different hypothetical physiological processes. The new models fitted the observations better than classical models. The bias of the classical models, i.e. overestimation of early observations and underestimation of late observations, could be reduced but not completely removed. Differences in the best-fitting model equations for each species indicated that, for the late spring phases (bud burst of Fagus sylvatica and Quercus robur), the photoperiod played a more dominant role than for early spring phases (bud burst of Betula pendula and Aesculus hippocastanum). Chilling only plays a subordinate role for spring bud burst compared to temperatures preceding this event in our data. The presented modeling approach allowed for a species-specific weighting of the dominant processes. The model results are in accordance with experimental findings that indicate an important role of day length in late spring BB. Potentials for model improvement are discussed.