In a modelling and simulation study the Utah Chill Unit and Growing Degree Hour submodels were tested. Based on litterature studies some of the latest knowledge and theories about temperature responses during the dormant period were incorporated into the models.
Tue Utah models were modified in two major variants:
I) Models involving a procedure which makes it possible to include the dynamic relationship between chilling and heat requirement.
II) Models involving a chain structure with each "link" representing a develop­mental phase through the dormancy period. Based on the teories proposed by Saure (1985), the temperature response of the system was adjusted in each phase in arder to simulate the change in the temperature response of the dormant tree.
Tue proposed models were used to predict the beginning of flowering inAlnus and Betula, which are closely related species, but shows great differences in chill- and heat requirement. Tue models were found to be very well correlated to the observed data. Tue phenological data were based on 15 years of pollen counts in Copenhagen.
Tue results of this study seem to support the model theories proposed by Saure. According to Saure the Chill Unit function is thought to represent the average response. However, in deep and late dormancy periods the response might be quite different.
A possible way to improve phenological temperature models might be to involve cultivar specific rutines which adjust the temperature fimetions according to the dormancy pattern caracteristic for the cultivar.