Modeling budburst in Coast Douglas-fir based on winter temperature and genotype

Background/Question/Methods

Models to predict the timing of budburst are needed to help predict some of the effects of climate change on ecosystems.  Many studies have demonstrated that overwintering buds sense cold temperatures (chilling) and warm temperatures (forcing), but the way different temperatures contribute towards chilling and forcing sums is not well understood for most species. In addition to environmental cues, budburst is under strong genetic control in some species.  Populations of Douglas-fir and other temperate species have adapted to a range of environments by altering their timing of budburst so that spring growth occurs when environmental conditions are optimal and the risk of damage is low.  Our goal was to develop a model to predict the timing of budburst in Douglas-fir based on winter temperature and genotype.  We conducted controlled experiments on 60 populations of coast Douglas-fir where trees were assigned to winter environments that had different levels of warming compared to the ambient environment.  Data from the experiment and scientific literature were used to develop chilling and forcing effectiveness functions to fit a “possibility line”, which indicates the limiting combination of chilling and forcing units that must be met before budburst is possible.

Results/Conclusions

Our chilling effectiveness function indicates that optimal chilling occurs between about 2° and 8° C but temperatures warmer and colder than this range are partially effective.  There were similarities between our function for Douglas-fir and functions developed for other species.  Our forcing effectiveness function indicates that temperatures greater than 22° C are optimal for forcing, but lower temperature are partially effective.  Some temperatures appear to be effective for both chilling and forcing.  Population of Douglas-fir differed in the amount of forcing required for budburst with a given amount of chilling; thus, different populations have different possibility lines.  Populations from dry, cold environments required less forcing for budburst than those from moist, wet environments.  Our final model can predict budburst of Douglas-fir based on seed-source environment and winter temperatures.  Our results indicate that with some warming budburst may happen earlier than in the past, but substantial warming will result in delayed budburst owing to a lack of chilling.  Experiments are now underway to determine if our chilling and forcing effectiveness functions can be adapted to predict budburst of other woody species.